A Three-Port Photovoltaic (PV) Micro-

Inverter with Power Decoupling Capability

Souhib Harb, Haibing Hu, Nasser Kutkut, Issa Batarseh, Z. John Shen
Department of Electrical Engineering and Computer Science
University of Central Florida
Orlando, FL 32826
ABSTRACT: This paper presents a new micro-inverter
topology that is intended for single-phase grid-connected
PV systems. The features of the proposed topology are:
(1) eliminating the double-frequency power ripple using
small film capacitor; (2) improving the maximum-
power-point tracking (MPPT) performance; (3) using
long life-time film capacitors, which will improve the
reliability of the inverter; and (4) requiring no
additional circuitry to manage the transformer leakage
energy.
Index terms- Power Decoupling, Single phase
Inverter, Photovoltaic, Microinverter
I. INTRODUCTION
Micro-Inverters connected to a single PV panel are
becoming the trend for the future of grid-connected PV
systems for a number of reasons including: (1) improved
energy harvest; (2) improved system efficiency; (3) lower
installation costs; (4) plug-N-play operation; (5) and
enhanced flexibility and modularity. Typically, the micro-
inverter is connected, and even attached, to a single PV
panel, which requires that the micro-inverter to have a life-
span matching the PV panels life-span, namely 25 years.
[1]. In single-phase grid-connected micro-inverters, the
MPPT provides constant output power from the PV panel
while the injected power to the grid is following a squared
sine wave.
P
0
(t) = I
0
I
0
sin
2
(
0
t) = P
0_u
+P
0
cc
(t)
(1)
As shown in Fig. 1. and assuming lossless inversion
process, the output power from the PV panel will be equal
to the average AC output power. The remaining time-
varying term P
o
ac
(t) will deteriorate the MPPT
performance; and as such, this pulsating power must be
handled by an energy storage device (power decoupling
capacitor), whose size is determined by:
v V
P
C
DC
in
D
~
0

(2)
where Pin is the rated power of PV panel, V
DC
is the DC
level voltage across the decoupling capacitor, v

is the
maximum allowable peak-to-peak ripple, and
0
is the line
frequency.

Usually, a capacitor is connected in parallel with the
PV panel, which results in a very large capacitance since
the allowable voltage ripple must be held to very low
values (<1%) to realize an efficient MPPT performance.
For a typical PV-Module with the following ratings: P
in
=
200W and v
C
D
(BC) = SSv and assuming a voltage ripple
v
C
D
= 2v, the minimum value of the decoupling capacitor
is 7.6mF in order to achieve a 98% PV utilization factor
[2]. Electrolytic capacitors are typically used due to their
large capacitance. However, the life-span of electrolytic
capacitors is relatively short compared with the life-span
of the PV panel, namely 1000~7000 hours at 105
o
C
operating temperature [3]. Some researchers have explored
various ways to reduce the size of the required capacitance
so as to allow for other longer life span capacitor
technologies, such as film capacitors, to be used [4-14].
These techniques have been presented in [15].
Some techniques employ power decoupling at the PV
side, such as the topologies proposed in [5] and [6]. The
decoupling capacitor is detached from the PV panel and
connected to the inverter through an auxiliary decoupling
circuit. This allows the DC level voltage and the voltage
ripple across the capacitor terminals to be increased
without affecting the MPPT performance; thus reducing
the required decoupling capacitance, (C
D
), where a small
film capacitor can be used. In addition to the losses
associated with the decoupling circuit, the double power
conversion in both topologies will increase the losses and
reduce the overall efficiency of the inverter. For the
Fig 1: Input and output power waveforms.
978-1-4244-8085-2/11/$26.00 2011 IEEE 203
topology in [5], a dissipative RCD clamp circuit is needed
to handle the transformer leakage energy problem, which
further increases the power losses. A modified version of
the topology in [5] was proposed in [6]; where the leakage
energy is stored in the decoupling capacitor through the
decoupling circuit and no dissipative snubber circuit is
needed. The decoupling circuit in [5] consists of one
switch and one diode while in [6] four switches and two
diodes are needed. Three port converters are proposed in
recent years to interface different renewable sources
[16].These topologies also can be modified to implement
power decoupling in single-stage micro-inverter
applications [17].In this paper, we propose a new topology
that employs power decoupling at the PV side; where the
transformer leakage energy is stored in the decoupling
capacitor with minimal added components. More details
will be given in next sections.
II. THE PROPOSED THREE-PORT MICRO
INVERTER
The proposed topology is presented in Fig. 2. The
power decoupling circuit consists of only one switch and
two diodes in addition to the decoupling capacitor. The
operation of this topology is divided into two main
modes: (1) charging mode; (2) discharging mode. The
decoupling capacitor will act as a secondary port. It will
work as a load in the first mode and as a source in the
second mode.
Grid
S1
S2
D1
D2
S4
S3
D3
D4
T
Cf
Lf
Cdc
CD
PV
Fig 2: The proposed Three-Port Micro-Inverter.
III.MODE OF OPERATION
The operation depends on the value of the output
power,
) (
0
t P
. From Fig. 1, the circuit will be operating in
mode-I (charging mode) when
in
P t P < ) (
0
. During this
mode, the extra energy will be stored in the decoupling
capacitor. When
in
P t P > ) (
0
, the circuit will be operating in
mode-II (discharging mode). The decoupling capacitor
will support the PV panel by discharging energy into the
transformer and transferring the required power to the
output (utility grid).
Mode-I
This mode is divided into three circuit modes, as
shown in Fig.3: (1) storing energy into the transformers
magnetizing inductance; (2) charging the decoupling
capacitor; (3) and transferring the power to the output.

(a)

(b)

(c)
Fig 3: Operation modes during charging mode (mode-I).
In the first mode S
1
is turned on (Figure 3a), the
magnetizing inductance will store the energy from the PV
panel, where the magnetizing current ramps up until it
reaches I
LP
, given in (3).
I
L
P
=
I
n
I
m

1
I
s
(3)
The peak value varies with time in order to keep the
input current (the current from the PV panel) constant.
The detailed derivation will be given later in this paper.
When the magnetizing current reaches the peak value, I
LP
,
S
1
is turned off. All switches are off during this mode
(Figure 3b). The transformer magnetizing energy will
discharge into the decoupling capacitor. The current will
keep discharging until it reaches the reflected value of the
secondary current at time (t
2
), which is given in (4).
i
L
m
(t
2
) = __
2I
0
I
s
I
0
I
m
_I
0
sin(
0
t) (4)
Note:
( ) t V
0 0
sin
can be considered as a constant value
given a very small switching cycle (T
s
) compared to the
output AC cycle (T
0
). At this point, one of the AC side
switches will be turned on (depending on the polarity of
the output voltage), Figure 3c. The power will be
transferred to the AC output side.
204
Using the power equation, I
L
P
can be derived as
P
n
=
1
2
I
L
P

1
I
s
I
n
+
1
2
[I
L
P
+ i
L
m
(t
2
)
2
I
s
I
n
I
s

(5)
It can be found that

1
I
s
=
I
m
I
L
P
I
n

(6)

2
I
s
=
I
m
(i
L
m
(t
2
) -I
L
P
)
I
n
- I
C
D
(7)
Substituting (6) and (7) in (5), then:
I
L
P
= __
I
n
I
C
D
[i
L
m
(t
2
)
2
- _
2P
n
I
s
I
m
] _
I
n
I
C
D
- 1__
(8)
Fig. 4 shows the operating waveforms during this
mode including the magnetizing current, the input and
secondary side current waveforms, and the switches
driving signal.
Fig 4: Inductor current, Input current, output current waveforms, S1, S2,
S3 gate signals for the two main operation modes.
Mode-II
This mode is also divided into three circuit modes, as
shown in Fig. 5. The first and the third modes are similar
to the previous modes. In the second mode, the
decoupling capacitor will be discharging its energy into
the transformers magnetizing inductance. The
transformers magnetizing current will be charged from
the PV panel by turning S
1
on, Fig. 5a. When the
magnetizing current reaches the first peak value, given in
(9), S
2
is turned on and the second mode starts at this
point of time, Fig. 5b.
I
L
P
= _
2P
n
I
s
I
m

(9)
The transformers magnetizing current will continue
charging from the decoupling capacitor until it reaches the
second peak value which is given in (4). At this time both
S
1
and S
2
are turned off (Fig. 5c) and the power is
transferred to the output side through one of the AC side
switches. Fig. 4 shows the magnetizing current, the input
and secondary side currents waveforms, and the switches
driving signals.

(a)

(b)

(c)
Fig 5: Operation modes during discharging mode (mode-II).
205
It can be noticed from the operation modes that the
leakage energy is stored in the decoupling capacitor
without using any additional circuit. This is one of the
advantages of the proposed topology. In [5], this leakage
energy will cause a spike on the decoupling switch at turn
off time when the energy is transferred to the AC side.
The modified topology in [6] solved this problem by
storing the leakage energy in the decoupling capacitor
through the flyback diodes. The advantages of the
proposed topology over the previous topologies in [5] [6]
can be summarized as follows:
1. No double power conversion, which results in
reduced power losses.
2. The transformer leakage energy is stored in the
decoupling capacitor through D
1
. This means
that there is no need for extra dissipative clamp
circuits. Again, this will reduce the power losses.
3. Fewer components are used in the decoupling
circuit, only one switch and two diodes, unlike
the topology in [6], where four switches and two
diodes are required.
IV. THE DECOUPLING CAPACITOR (C
D
)
The formula that governs the decoupling capacitor
(C
D
) value will be derived in this section. During sub-
mode-3 in both main modes, the energy that has been
stored in the magnetizing inductance will be transferred to
the secondary side. Wherefore, the voltage across the
decoupling capacitor must be greater than the stress
voltage across S
1
, which is given in (10).
I
S
1
= I
n
+
n
1
n
2
I
0
sin(2
0
t)
(10)
Then, the minimum voltage across C
D
must satisfy the
following condition (11).
I
C
D
mn
I
n
+
n
1
n
2
I
0

(11)
But, the minimum voltage across C
D
is given in (12).
I
C
D
mn
= I
C
D
(C) -
I
C
D
2

(12)
Using (11) and (12), the relationship in (13) can be
derived. By choosing a certain DC-level across the C
D
,
the maximum allowed voltage ripple can be found.
I
C
D
2 _I
C
D
(C) -I
n
-
n
1
n
2
I
0
]
(13)
It is worth to mention that when we choose the DC-
level voltage across the decoupling capacitor we should
take in the consideration the stresses on the power devices
at the input side (S
1
, S
2
, D
1
, and D
2
). Now, for a certain
system parameters, we can choose the appropriate value
for the decoupling capacitor. Consider a 100W system
with 35V as an input voltage with a grid voltage and
frequency of 110V
rms
and 60Hz, respectively. Let us
choose a DC-level across the decoupling capacitor of
120V. Then the ripple should be less than 107V.
Assuming a value of 100V and using (2), 22.1F of
decoupling capacitance is needed.
V. SIMULATION RESULTS
The PSIM software is used to simulate the proposed
topology. Table 1 lists the components values that were
used in the simulation.
Table 1: Proposed topology component values.

Fig. 6 shows the simulation results. In Fig. 6, the
magnetizing inductance, input, and secondary currents
waveforms for one complete grid cycle are shown.
Magnetizing inductance
current, iLm(t)
Input current, iin(t)
Secondary side
current, i2(t)
Fig 6: Simulation results of the proposed topology.
Fig. 7 and Fig. 8 show an expanded view of the circuit
operation in mode-I and mode-II, respectively.

Fig 7: Expanded view for the simulation results for mode-I.
Input Power (P
in
) 100 W Input Voltage (V
in
) 35 V
Magnetizing
Inductance (L
m
)
7H Turns Ration (N) 5
Decoupling
Capacitance (C
D
)
40F C
f
1F
Output Resistor (R
o
) 120 L
f
1mH
206

Fig 8: Expanded view for the simulation results for mode-II.
Fig. 9 shows the output current, output voltage, I
C
D
,
the stress voltage across S
1
and I
C
D
, and finally the input
and output power waveforms.

Fig 9: Output current and voltage, VCD, the stress voltage across S1 and
VCD, and finally the input and output power waveforms.
VI. EXPERIMENTAL RESULTS
A prototype was set up to verify the validity of the
proposed topology. To simplify the verification, the
power circuit for the topology, shown in Fig.10, with an
open loop control strategy is employed. The key
parameters are listed in table 2.
Table 2: The experimental parameters' values.
Input voltage 20V
Switching frequency 100KHz
Magnetizing inductance 7.6uH
Transformer turn ratio 1:1
R0 40
CD 50uF
Cr 1uF
Fig.10: Power stage circuit for verification purposes.
Fig.11 shows the driver signals in Mode I operation,
during which the switch S
2
is always off. When switch S
1

turns off, the energy stored in the magnetizing inductance
is transferred to the capacitor C
D
and switch S
3
turns on
when the current falls to certain level. The current
waveforms, both in transformer primary side and
secondary side, are depicted in Fig.12. Fig.13 shows the
driver signals in Mode II operation, during which the
switch S
3
is always on. The energy stored in the capacitor
increases the magnetizing current by turning on the switch
S
2
as shown in Fig.14. All these driver signals are
calculated and generated by STM32 ARM based on the
reference current, input DC voltage and capacitor voltage.

Fig.11: Driver signals in Mode 1

Fig.12: the current waveforms in both primary and secondary sides
S
1
S
3

S
2

Current at primary side
Current at secondary side
207

Fig 13: Driver signals in Mode 2

Fig.14: Current waveforms in Mode II

Fig. 15: output voltage and voltage across the decoupling capacitor
The output voltage as shown in Fig.15 is rectified
sinsoidal waveform. Some slight distortions can be
noticed during mode transitions due to nonideal factors
such existance of leakage inductance, switch transients
and calculation errors in the MCU. The voltage across the
decoupling capacitor fluctates according to charging and
discharging process. As shown in Fig. 15, at the point
where the operation mode switches from mode I to mode
II, the voltage across the capacitor is charged to maximum
and the minimal voltage occurs at mode transition point
from mode II to mode I.
VII. CONCLUSION
A new micro-inverter topology is presented. It is
primarily intended for the AC-Module PV systems. The
proposed topology employs new power decoupling
technique where a small film capacitor can be used
instead of electrolytic capacitor. Hence, it will have a
long life-span comparable to the PV panel. The
transformer leakage energy is handled by the decoupling
circuit itself so there is no need for additional dissipative
circuits; which leads to reduced power losses and
improved efficiency.
REFERENCES
[1] S.B.Kjaer, J.H.Pedersen, F. Blaabjerg. A Review of Single-
Phase Grid-connected Inverters for Photovoltaic Modules. IEEE Trans.
Industry Applica. Vol.41 no.5, pp.1292-1306, Sept. 2005.
[2] S. B. Kjaer, Design and control of an inverter for photovoltaic
applications, Ph.D. dissertation, Inst. Energy Technol., Aalborg
University, Aalborg East, Denmark, 2004/2005.
[3] www.CDE.com// Type 381EL 1050C Ultra-Long Lift Snap-In,
Aluminum.
[4] A.C. Kyritsis, N. P. Papanikolaou, E. C. Tatakis. A novel
Parallel Active Filter for Current Pulsation Smoothing on Single Stage
Grid connected AC-PV Modules, in Proc. European Conf. Power
Electronics and Applications (EPE), 2007.
[5] T. Shimizu, K. Wada, and N. Nakamura, Flyback-type single-
phase utility interactive inverter with power pulsation decoupling on the
dc input for an ac photovoltaic module system, IEEE Trans. Power
Electron., vol. 21, no. 5, pp. 12641272, Sep. 2006.
[6] S. B. Kjaer and F. Blaabjerg, "Design optimization of a single
phase inverter for photovoltaic applications," in Proc. IEEE Power
Electronics Specialists Conference, 2003, vol. 3, pp. 1183-1190.
[7] G. H. Tan, J. Z. Wang, and Y. C. Ji, Soft-switching flyback
inverter with enhanced power decoupling for photovoltaic applications,
IET Trans. Elect. Power Appl., vol. 1, no. 2, pp. 264274, Mar. 2007.
[8] P. Enjeti and W. Shireen, A new technique to reject DC link
voltage ripple for inverters operating on programmed PWM
waveforms, IEEE Trans. on Power Electronics, VOL. 7, NO. 1, Jan.
1992.pp.171-180.
[9] T. Brekken. N. Mohan, C. Henze, & L.R. Moumneh, "Utility-
connected power converter for maximizing power transfer from a
photovoltaic source while drawing ripple-free current," IEEE PESC,
2002.pp.1518-1522.
[10] N. A. Ninad, L. Lopes, A Low Power single-Phase Utility
Interactive Inverter for Residential PV Generation with Small DC Link
Capacitor
[11] P. T. Krein and R. S. Balog, "Cost-effective hundred-year life for
single-phase inverters and rectifiers in solar and LED lighting
applications based on minimum capacitance requirements and a ripple
power port," IEEE Applied Power Electronics Conference, 2009.
[12] Al-Atrash, H.; Tian, F.; Batarseh, I., Tri-Modal Half-Bridge
Converter Topology for Three-Port Interface, IEEE Trans. Power
Electronics, vol. 22, pp. 341 - 345, January 2007.
[13] Q. Li, P. Wolfs, and S. Senini, A hard switched high frequency
link converter with constant power output for photovoltaic applications,
in Proc. Australasian Univ. Power Eng. Conf., 2002.
[14] C. Bush, B. Wang. A Single-Phase Current Source Solar
Inverter with Reduced-Size DC Link IEEE Energy conversion congress
and Exposition, San Jose,CA, 2009,pp.54-59.
[15] H. Hu, S. Harb, N. Kutkut, I. Batarseh, Z. J. Shen, Power
Decoupling Techniques for Micro-inverters in PV Systems-a Review,
IEEE Energy conversion congress and Exposition, 2010.
[16] Zhijun Qian; Abdel-Rahman, O.; Al-Atrash, H.; Batarseh, I.; ,
"Modeling and Control of Three-Port DC/DC Converter Interface for
Satellite Applications," Power Electronics, IEEE Transactions on ,
vol.25, no.3, pp.637-649, March 2010
[17] Zhijun Qian; Abdel-Rahman, O.; Hu, H.; Batarseh, I.; , " An
Integrated Three-port Inverter for Stand-alone PV Applications," Energy
Conversion Congress and Exposition, 2010. ECCE 2010. IEEE , vol.,
no., pp.1471-1478, 12-16 Sept. 2010

S
1

S
2

S
3

Output voltage
Voltage across Cd
Current at primary
id
Current at secondary
id
208