Power Decoupling Method for Single Stage Current

Fed Flyback Micro Inverter

M.SeyyedHosseini1 , A.Yazdian-Varjani2 , M.Mohamadian3
Faculty of Electrical and Computer Engineering
Tarbiat Modares University
Tehran, Iran
1
m.seyyedhosseini@modares.ac.ir
2
yazdian@modares.ac.ir
3
mohamadian@modares.ac.ir

Abstract—Among different micro inverter topologies, single ipv ig

stage micro inverters duo to their lower component, low cost +
and simple control structure are paid more attentions. One of Solar C vg
+
grid
vpv
DC-AC −
the biggest problem in single phase micro inverter is power ripple Module −
in twice frequency of line. This phenomena causes voltage and
current ripple in the side of PV so that designing MPPT controller
becomes very hard. To eliminate power ripple (power decoupling)
Fig. 1. Grid connected micro inverter
in conventional micro inverters, large electrolytic capacitors are
used in parallel of PVs. Drawbacks of electrolytic capacitors
are large ESR and lower life cycle in comparison of other type
1 1
of capacitors. In this paper, a power decoupling method for Po = v g i g = Vg Ig − Vg Ig cos(2ωg t) = Ppv −Ppv cos(2ωg t)
Single stage flyback based current fed micro inverter is presented. 2 2
By this method, the required capacitance for power decoupling (3)
and current stress of semiconductors are decreased which film As seen in equation (3), grid power demand in this case
capacitors can be used instead of electrolytic. The effectiveness consists of two sub power term. Pulsating power, that oscillates
of this method is confirmed for 100w micro inverter by PSIM at twice of the line frequency and constant average power.
simulation. Fig.2 shows the grid power demand and generated solar power.
Index Terms—micro inverter, power decoupling
The solar control system must ensure that the power from solar
module is in maximum power point (MPP). Since maximum
I. I NTRODUCTION
power absorbed from solar module is equal to average grid
Use of alternative energy sources instead of fossil fuels help power demand so there is always a deficiency and surplus
reduce the pollution level. Green energies such as solar, wind between grid power demand and generated solar power.
and etc, are among the best alternative sources for fossil fuels. In order to keep the system stable while the energy is
Among them, Solar energy are uptrend in study cases due flowing back and forth, a large storage component must
to health, environmental and accessibility features. Ability to be used. Inductors and capacitors are the the two storage
convert solar irradiation into electric power directly, has made element options, but usually capacitors are used. Decoupling
solar cells a great choice in renewable energy [1]. capacitors eliminate the effect of pulsating power on solar
Inverters have essential role in solar systems for converting module terminals. Location of decoupling capacitor is different
solar power to energy form, accessible to public. Centralized, in various topologies. It can be in parallel with the solar
string and micro inverters are three types of inverters to convert
solar DC energy to alternating AC electric energy. Micro
inverters are paid more attentions because of great features
such as: harvesting maximum available solar energy, high Po
efficiency, modularity and independent operation in large solar Release
Ppv
system [2].
Fig.1 shows the block diagram of a conventional grid con-
nected micro inverter. In order to achieve achieve unity power Store Store

factor, grid current and grid voltage must be in phase.

vg = Vg sin(ωg t) (1)
π π 3π
4 2 4
ig = Ig sin(ωg t) (2) ωt

978-1-7281-5849-5/20/$31.00 ©2020 IEEE Fig. 2. Energy flow in grid connected MI

Authorized licensed use limited to: Fondren Library Rice University. Downloaded on May 17,2020 at 06:03:42 UTC from IEEE Xplore. Restrictions apply.
 D4 S3
Lf
Cf
PV C1
D5 S4
S1
Fig. 3. Basic current fed topology micro inverter
module or within inverter stages. Single phase operation of
inverters causes power pulsation. This pulsation causes solar
module voltage and current disturbance [3].
The required capacitance for power balancing is:
Ppv
C=
ωg vpv δvpv
The required capacitance depends on PV power (Ppv ), line
frequency (ωg ), PV voltage (vpv ) and allowed voltage ripple
on PV (δvpv ). For example a 100w micro inverter connected
to a 30v PV module with allowed 1v voltage ripple, requires
approximately a 10mF capacitor which is large and low
efficient.
II. BASIC CURRENT FED FLYBACK MICRO INVERTER
Fig.3 shows the basic structure of flyback current fed micro
inverter. This one stage flyback based structure can transfer
power from solar module to grid. This inverter by help of
two secondary windings transfer solar energy from the solar
module to grid. In every half cycle of grid, one of the
secondary windings plays the role of transferring energy to
the grid. So every winding has a unidirectional switch which
controlled by the mains voltage half cycle detection unit or
PLL system. The low-pass filter (Lf ,Cf ) in the grid side
is placed in order to improve harmonic distortion due to
discontinuous current operation of this micro inverter. As low-
pass filter (Lf ,Cf ) in micro inverter due to low power range
has negligible voltage drop, discussing micro inverter acts like
a dc to dc flyback converter which its output voltage can be
controlled [4].
Operation of grid connected flyback micro inverter is based
on shaping peak current in each switching cycle to follow the
sinusoidal grids voltage. To establish this idea switch S1 must
be trigger by sinusoidal duty cycle pattern [5].
d(t) = dp sin(ωg t)
dp is the duty cycle of switching cycle at ωg t =
π/2 of grid voltage. As dp increases, power absorbed
from module and peak current increases. This micro in-
verter can operate in DCM(discontinuous-conduction mode),
BCM(boundry-conduction mode) and also CCM(continuous-
conduction mode) that discussed in [6].
The drawbacks of flyback micro inverter topology are high
Authorized licensed use limited to: Fondren Library Rice University. Downloaded on May 17,2020 at 06:03:42 UTC from IEEE Xplore. Restrictions apply.
D4 S3
Lf
D1 n1 n3
S2 n2 Cf
PV C1 D2
C2
D5 S4
n4
D3 S1
Fig. 4. Proposed topology
current stress of switches and large decoupling capacitor (C1 ).
Current injected to (Lf ,Cf ) filter has sinusoidal envelop and
average of this current pass to grid. So the maximum peak
current is high. The large current stress implicit in a flyback
working in DCM would reduce the conversion efficiency,
(4) which is a critical requirement particularly in an AC module
application. Because of power pulsating on solar module,
large capacitor required to eliminate voltage ripple in input
of circuit.
Currently, some methods have been introduced to reduce the
amount of decoupling capacitor especially for flyback type
micro inverters, so that life time of whole micro inverter can be
improved by using more resisting capacitor like film capacitors
[7]- [11].
III. C IRCUIT CONFIGURATION AND OPERATION PRINCIPLE
OF PROPOSED TOPOLOGY
The proposed topology by adding some components to the
basic current fed micro inverter topology can decouple power
pulsating. Current from solar module always maintain constant
so solar module parallel capacitance required is low.
The proposed topology with power decoupling
capacitor(C2 ) is shown in Fig.4. Every half cycle of
main power voltage is divided into three part according
to power difference between grid and solar module [0, π4 ],
[ π4 , 3π 3π
4 ] and [ 4 ,π] as shown in Fig.2. In this topology,
inverter always absorbs constant power from input in each
cycle. So in each cycle whole solar power present in primary
magnetizing inductor of transformer. Thus, energy flowing is
different in each triple period of grid.
In [0, π4 ] period of grid voltage showing in Fig.2 the grid
power demand is lower than solar power, so grid power
demand transferred by transformer and further energy is
stored in decoupling capacitor. In [ π4 , 3π4 ] period, grid power
demand is larger than solar power, so initially, total solar
(5) power transferred by transformer to the grid and extra demand
is supplied by energy stored in capacitor in [0, π4 ] and [ 3π 4 ,π]
period. In [ 3π4 ,π] similar to first period ([0, π
4 ]) the grid power
demand is lower than solar power, so grid power demand
transferred by transformer and further energy is stored in
decoupling capacitor.
Average current in every cycle must be equal to in phase
sinusoidal current of grid. In [ π4 , 3π
4 ] period of grid voltage
due to existence of two different current source(current from
solar module and current from decoupling capacitor), average
of two current must be equal to sinusoidal current so current S3
iin D4 if ilter
envelope in this part is flat. This property causes lower current
stress in semiconductors. Also this topology only required n1 n3
Lf
D1
two capacitors, capacitor in parallel of solar module(C1 ) S2 n2 Cf

perform energy storage role for each cycle and its value
PV C1 D2
is determined by equation (4) only ωg substituted by ωs . C2 D5 S4

Because ωs is much higher than ωg so this capacitor is much ic2

n4

lower than capacitor in conventional micro inverter. Another D3 S1

capacitor (C2 ) used to decouple power. Power oscillation

at twice line frequency appear on this capacitor. Voltage
oscillation on C2 can be high to obtain lower capacitance (a) State 1
S3
according to equation (4). iin D4 if ilter

The proposed topology divided in 5 states in [0, π4 ],[ π4 , 3π

4 ] and
Lf

[ 3π
4 ,π] period of grid voltage in one half cycle. In positive
D1
S2 n2
n1 n3
Cf
and negative half cycle of grid all 5 states are confirmed.
Positive and negative half cycle only differ in controlling of PV C1 D2
C2 D5 S4
ac switches (S3 and S4 ). In positive half cycle switch S3 is n4
ic2
controlled and in negative half cycle switch S4 is controlled. D3 S1

IV. O PERATING STATES

Fig.5 shows all operational states of proposed topology. (b) State 2
States are described in detail in this section. iin D4 S3 if ilter

A. State 1 Lf
D1 n1 n3
In Fig.6, at the beginning of each cycle switch S1 turns on S2 n2 Cf

until dTs and gets constant power from module and stores this
PV C1 D2
energy in transformer magnetizing inductance(Lm1 ). Input C2 D5 S4

peak current after dTs determined by: ic2

n4

D3 S1
Vpv dTs
iinpeak = (6)
Lm1
For calculating power absorbed from solar module, average of (c) State 3
input current in switching period is calculated. iin D4 S3 if ilter

2
Vpv d Ts Lf
< iin >= (7) D1 n1 n3
2Lm1 S2 n2 Cf

2 2
Vpv d Ts PV C1 D2
Ppv =< iin > Vpv = (8) C2
2Lm1
D5 S4
n4
ic2
So switch S1 always is driven by constant duty cycle(d). dTs D3 S1

determined by power demand from solar module.


2Ppv Lm1
d= 2T
(9) (d) State 4
Vpv s iin D4 S3 if ilter

B. State 2 Lf
D1 n1 n3
In this state, all semiconductors are off. Part of energy S2 n2 Cf

stored in magnetizing inductor(Lm1 ) release in decoupling

PV C1 D2
capacitor. Amount of energy released in decoupling capacitor C2 D5 S4

is controlled by start of next state. ic2

n4

mc2 is the slope of current injected to decoupling capacitor. D3 S1

vc2
mc2 = − (10)
Lm1
(e) State 5
After dc Ts , decoupling capacitor current decreases from
iinpeak to ic2 . Fig. 5. Proposed topology operating states

ic2 = iinpeak + mc2 dc Ts (11)

Authorized licensed use limited to: Fondren Library Rice University. Downloaded on May 17,2020 at 06:03:42 UTC from IEEE Xplore. Restrictions apply.
Average current injected to decoupling capacitor is determined T T T
iin
by: peak

iinpeak + ic2 Iin

< ic2 >= (12)
2Ts
ic2
Average power injected to decoupling capacitor is determined
IC2
by (13). This power must be satisfied by grid extra power
demand. if ilter
ic2max
peak

2iinpeak − vc2 dc Ts
Lm1
If ilter

< pc >= vc2 < ic2 >= vc2 (13)

2Ts
S1
C. State 3
In third state, switch S3 and diode D4 turn on and energy
S2
stored in magnetizing inductor transferred to (Cf ,Lf ) filter by
transformer. Considering the fact that in low power inverters,
S3 , S 4
phase different between grid voltage and filter voltage is
negligible, it can be assumed that the voltage across Cf is d dc d o d do dc d dc d o ωt
same as grid voltage. do is specified in Fig.6 and is the time

(state1)
(state2)
(state3)

(state1)

(state3)

(state4)
(state5)

(state1)
(state2)
(state3)
π 3π π
4 4
when if ilter falls to zero. Due to current slope in magnetizing
inductor:
vgrid n1 2
if ilterpeak = ( ) do (14) Fig. 6. Waveforms of parameters
fs Lm1 n3
if ilterpeak and ic2 are related by transformer turn ratio:
if ilterpeak n1
= (15) In this state switch S1 and diode D2 turn on and energy
ic2 n3
stored in decoupling capacitor run in magnetizing inductor
Grid power demand must be equal to average power trans- (Lm2 ). This energy flow continues until extra grid power
ferred by transformer to (Cf ,Lf ) filter. demand is satisfied. dc is specified in Fig.6. dc Ts is duration
Pgrid = Ppv − Ppv cos(2ωg t) (16) of state 4 that last until specified power is absorbed from
decoupling capacitor.
if ilterpeak do By turning on switch S1 current ramps up to ic2max .
< if ilter >= (17)
2
vc2 
< if ilter > vgrid = Ppv − Ppv cos(2ωg t) (18) ic2max = d T (22)
Lm2 c
By manipulating (11), (14), (15), (16) and (17) in (18) dc is Average current from decoupling capacitor is determined by:
calculated as below:
 ic2max dc
fs Lm1 n3 2Ppv (1 − cos(2ωg t)) < ic2 >= (23)
dc = [iinpeak − )] (19) 2
vc2 n1 fs Lm1
Power from decoupling capacitor must compensate extra grid
dc Ts is the duration of second state. After (d + dc )Ts switch power demand.
S3 turns on.
< ic2 > vc2 = −Ppv cos(2ωg t) (24)
D. State 4
State 4 only occurs in [ π4 , 3π By substituting (22) and (23) in (24) dc is calculated:
4 ] period of grid voltage and
right after completely solar module energy transferred by 
 −2fs Lm2 Ppv cos(2ωg t)
transformer. As mentioned in previous State, do is the time dc = (25)
vc2
last if ilter falls to zero.
iinpeak fs Lm1 E. State 5
do = (20)
vgrid By turning off switch S2 in this state, energy stored in
magnetizing inductor (Lm2 ) transferred to output by trans-
Maximum of do occurs in π4 degree of grid voltage. After
former. Energy transferred from state 1 in [ π4 , 3π
4 ] and energy
domax all solar module power transferred by transformer and
transferred from decoupling capacitor add up and supply grid
state 4 can be started.
√ power demand. Process of adding and current fragmentation
2iinpeak fs Lm1 lead to decreasing current peaks and decreasing semiconductor
domax = (21)
vgridpeak current stress.

Authorized licensed use limited to: Fondren Library Rice University. Downloaded on May 17,2020 at 06:03:42 UTC from IEEE Xplore. Restrictions apply.
 Table. I. Components value
Sensors
ܸ௚௥௜ௗ ܸ௖ ܸ௣௩ ‫ܫ‬௣௩ parameter value
vpv 30v
vgrid 220v
MPPT fs 50kHz
PLL C1 40uF
Eq(25) Eq(19)
‫כ‬
ܲ௣௩ C2 20uF
߱‫ݐ‬௚௥௜ௗ Cf 1uF
݀ Lf 1mH
݀Ԣ௖ ݀௖ ݅௜௡௣௘௔௞ Eq(9)
Half-cycle Lm1 20uH
detection ݀௢௠௔௫
݀ n1 : n2 : n3 : n 4 1:2:3:3

Period ܵଶ
detection

ߨ ͵ߨ ߨ ͵ߨ
Ͳǡ ܽ݊݀ ǡߨ ǡ
Ͷ Ͷ Ͷ Ͷ

ߨ ͵ߨ
Ͳǡ ܽ݊݀ ǡߨ
Ͷ Ͷ

ߨ ͵ߨ
ǡ
Ͷ Ͷ
ܵଵ
ܵଶ
ܵଷǡସ

Fig. 7. Block diagram of controller and switching signals

V. C ONTROLLER AND SWITCHING SIGNALS

Fig.7 shows block diagram of the proposed inverter. It
includes sensors for measuring PV voltage(Vpv ), decoupling
capacitor voltage(VC2 ), grid voltage(Vgrid ) and PV current
(Ipv ). Phase lock loop(PLL) block, MPPT controller, half cycle
and period detection are main blocks in this control strategy.
The control system is programmed to get the maximum power
available from the PV and inject it to the grid at maximum
power factor.
Period detection block also determine three specified period.
[0, π4 ] and [ 3π
4 ,π] period of grid voltage where power from
solar module is higher than grid power demand and [ π4 , 3π 4 ]
period of grid voltage where power from solar module is
lower than grid power demand. According to Fig.6 showing
waveforms switch signals are created by logical operation on
previous formula that calculate d, dc , do and dc .
VI. S IMULATION
A 100W inverter was designed and simulated in PSIM
software according to design procedure in previous sections.
Component values obtained from the Table 1.
Transformer play a major role in this topology. It is assumed
that leakage inductance of transformer is negligible. Current
that flow to (Cf ,Lf ) filter, current that injected to grid and
decoupling capacitor current are shown in Fig.8. As mentioned
in previous section, current flow to filter has two current part,
for this reason current envelop is flat in [ π4 , 3π
4 ] period of grid
voltage but the average of current every cycle in this period
is sinusoidal.
Current injected to grid is approximately in phase to grid
voltage. Current THD is lower than 5%. As (Cf ,Lf ) filter
frequency decreases by increasing Cf and Lf , the THD gets
lower but phase lag increases.
Fig. 8. Current waveforms of filter current, grid current and
decoupling capacitor
Fig.9 shows the voltage across decoupling capacitor and
grid voltage. As mentioned before, power pulsating on decou-
pling capacitor has twice line frequency variation. As seen in
Fig.9 voltage across decoupling capacitor has 110V peak to
peak and 100Hz frequency.
Fig.10 shows the input, decoupling capacitor and filter
current of micro inverter in [0, π4 ] period of grid voltage. Fig.11
shows the input, decoupling capacitor and filter current of
micro inverter in period [ π4 , 3π4 ] period of grid voltage. As
mentioned in states definition and waveform Fig.6, simulation
results confirms the theory.
VII. C ONCLUSION
In this paper a new method to decouple power, using
unidirectional switches, for a single stage current fed flyback
inverter was introduced for grid connected micro inverters.
Decoupling capacitor significantly decreased by this method so
large electrolytic capacitors can be replaced by film capacitors
witch have higher life cycle and lower ESR. Also current
stress in transformer secondary is decreased due to average
current operation of inverter. The operation of this circuit is
explained and simulation results are provided to demonstrate
its effectiveness.

Authorized licensed use limited to: Fondren Library Rice University. Downloaded on May 17,2020 at 06:03:42 UTC from IEEE Xplore. Restrictions apply.
 Iin
Vc
20
200

15
180

10
160

5
140

0
120
Ic

100
0

80 -2

-4
Vgrid
-6
400
-8

-10
200
I_filter

0 6

4
-200

2
-400

0.06 0.07 0.08 0.09 0.1 0

Time (s) 0.083 0.084 0.085 0.086 0.087
Time (s)

Fig. 9. Voltage of decoupling capacitor and grid voltage Fig. 11. Waveforms of input, decoupling capacitor and filter
current in [ π4 , 3π
4 ] period
Iin

20

International Conference on Power Electronics and Drive Systems. IEEE

15
PEDS 2001-Indonesia. Proceedings (Cat. No. 01TH8594). Vol. 2. IEEE,
10 2001.
5
[5] A. Ch, E. C. Tatakis, and N. P. Papanikolaou. ”Optimum design of
the current-source flyback inverter for decentralized grid-connected
0
photovoltaic systems.” IEEE transactions on energy conversion 23.1
Ic
(2008): 281-293.
20
[6] Y. Li & R. Oruganti (2009) A Flyback-Ccm Inverter Scheme For
15 Photovoltaic AC Module Application, Australian Journal of Electrical
10
and Electronics Engineering, 6:3, 301-309
[7] T. Shimizu, K. Wada, N. Nakamura, ”Flyback-Type Single-Phase Utility
5
Interactive Inverter With Power Pulsation Decoupling on the DC Input
0 for an AC Photovoltaic Module System,” IEEE Transactions on Power
I_filter
Electronics , vol.21, no.5, pp.1264-1272, Sept. 2006
[8] A.C. Kyritsis, N.P. Papanicolaou,; E.C. Tatakis, ”A novel Parallel
6
Active Filter for Current Pulsation Smoothing on single stage grid-
4
connected AC-PV modules,” in Power Electronics and Applications,
2007 European Conference on , vol., no., pp.1-10, 2-5 Sept. 2007
2
[9] A.C. Kyritsis, N.P. Papanikolaou, E.C. Tatakis, ”Enhanced Current Pul-
sation Smoothing Parallel Active Filter for single stage grid-connected
0 AC-PV modules,” in Power Electronics and Motion Control Conference,
0.0805 0.081 0.0815 0.082 0.0825
Time (s) 2008, vol., no., pp.1287-1292, 1-3 Sept. 2008
[10] Hu, Haibing, et al. ”A review of power decoupling techniques for
microinverters with three different decoupling capacitor locations in PV
Fig. 10. Waveforms of input, decoupling capacitor and filter systems.” IEEE Transactions on Power Electronics 28.6 (2012): 2711-
current in [0, π4 ] period 2726.
[11] Zare, Mohammad Hadi, Mustafa Mohamadian, and Reza Beiranvand.
”A single-phase grid-connected photovoltaic inverter based on a three-
switch three-port flyback with series power decoupling circuit.” IEEE
R EFERENCES Transactions on Industrial Electronics 64.3 (2016): 2062-2071.

[1] Chiu, Huang-Jen, et al. ”A module-integrated isolated solar microin-

verter.” IEEE Transactions on Industrial Electronics 60.2 (2012): 781-
788.
[2] H. Hu, Q. Zhang, X. Fang, Z. J. Shen, and I. Batarseh, ”A single stage
micro-inverter based on a three-port flyback with power decoupling
capability,” in Proc. IEEE ECCE, Sep. 2011, pp. 1411-1416.
[3] Kyritsis, A. C., N. P. Papanikolaou, and E. C. Tatakis. ”A novel
parallel active filter for current pulsation smoothing on single stage
grid-connected AC-PV modules.” 2007 European Conference on Power
Electronics and Applications. IEEE, 2007.
[4] Konishi, Y., et al. ”Utility-interactive high-frequency flyback transformer
linked solar power conditioner for renewal energy utilizations.” 4th IEEE

Authorized licensed use limited to: Fondren Library Rice University. Downloaded on May 17,2020 at 06:03:42 UTC from IEEE Xplore. Restrictions apply.
 A High Efficiency Flyback Micro-Inverter with A
New Adaptive Snubber for Photovoltaic Applications
Mohammad Ali Rezaei, Kui-Jun Lee, Alex Q. Huang, Fellow, IEEE
FREEDM Systems Center, North Carolina State University
Raleigh, NC 27606, USA
Abstract— Based on the hybrid operation of interleaved
flyback micro-inverter in Discontinuous and Boundary
Conduction Modes (DCM and BCM), a novel adaptive
snubber is proposed in this paper. The proposed snubber
limits the drain-to-source voltage overshoot of the
flyback’s main switch during the turn-off process,
enabling the use of lower voltage MOSFETs. It also
recovers the stored energy in the leakage inductance of the
flyback transformer and provides soft switching for the
main flyback switch by limiting the rising slope of the
MOSFET voltage during the turn off process resulting in
higher efficiency. Exploiting the natural resonant of the
flyback converter in BCM the adopted controller provides
ZVS and ZCS for the main switch during the BCM
operation. The operation of the flyback micro-inverter
with associated controllers is analytically studied, and
considerations for an optimum design aiming to higher
efficiency are presented. Performance of the flyback
micro-inverter with the proposed adaptive snubber and
the corresponding controllers are experimentally verified
based on a 250W interleaved flyback micro-inverter
hardware setup.
Keywords— Flyback, High Efficiency, Micro-Inverter, Snubber
I. INTRODUCTION
Photovoltaic (PV) micro-inverters have gained a
significant attention for grid-connected PV system
applications during the past few years because of improved
energy harvest, friendly “Plug-N-Play” operation, and
enhanced modularity and flexibility [1]-[3]. Various inverter
topologies for PV micro-inverters applications have been
introduced in the literature that perform the maximum power
point tracking (MPPT) of PV module, high step-up voltage
amplification, output current shaping, and galvanic isolation
[1], [3]. Among them, the flyback based micro-inverter is one
of the most attractive solutions due to its simple structure and
control and also inherent galvanic isolation.
Fig. 1 illustrates a flyback micro-inverter, which consists
of decoupling capacitor(s), interleaved flyback converter,
unfolding bridge, and CL filter. The unfolding bridge is
switched at line frequency by a simple square-wave control,
generating a rectified sinusoidal waveform at the dc-link
This work was supported in part by the U.S. Department of Energy under
award number DE-EE-0006036 and made use of ERC shared facilities
supported by the National Science Foundation under Award Number EEC-
0812121.
This work has been accepted for publication in IEEE Transactions on
Power Electronics.
978-1-4673-7151-3/15/$31.00 ©2015 IEEE
between the interleaved flyback converter and unfolding
bridge. The decoupling capacitor, which consists of several
parallel connected electrolytic capacitors, maintains the power
balance between the constant input power and variable output
power oscillating at double-line-frequency. All the other
functionalities required in PV micro-inverter are performed by
the flyback converter. Therefore, the flyback converter has
been widely scrutinized to improve its performance in terms
of efficiency, reliability, and cost.
In order to improve efficiency and power density, both the
discontinuous conduction mode (DCM) and the boundary
conduction mode (BCM) operations have been
comprehensively analyzed and a hybrid switching strategy has
been adopted [4, 5]. To further improve the weighted
efficiency of the micro-inverter system, synchronous
rectification at the high voltage side of the flyback inverter is
employed in [6], which leads to slight improvement in the
overall efficiency of the flyback micro-inverter system.
A non-complementary active clamp control method is
proposed to achieve high efficiency both for full-load and
light-load condition by reducing a circulating energy [7].
Furthermore, appropriate design parameters under DCM
operation have been specified to obtain maximum weighted
efficiency through analytical loss calculation [8] and adaptive
active clamping and phase control method [9] and hybrid
control strategy combing two-phase and one-phase DCM
control [10] have been proposed to improve weighted
efficiency according to the output power of a PV module.
Using a center-tapped transformer and bidirectional switched
on the grid-side, a turn-on ZVS approach is proposed in [11]
which utilizes snubber to maintain the voltage spike within
limits.
In [12], an open-loop control based on the feedback of the
primary current in the continuous conduction mode (CCM)
operation has been proposed to accommodate a right half plane
(RHP) zero in the control to output current transfer function,
thus it showed an efficiency improvement compared to the
DCM scheme. A through dynamic modeling and control of an
interleaved flyback micro-inverter along with a two-step
controller design based on realistic fourth-order system
modeling including the dynamics of the output CL filter is
presented in [13] which aims to compensate the instability
issue of the converter working in CCM. Also, in order to
reduce the cost and complexity of the flyback micro-inverter
system, the idea of multiple-integrated converter modules
sharing a single unfolding bridge is presented in [14] based on
the flyback micro-inverter operating in CCM.
A procedure for selecting the decoupling capacitors in the
flyback micro-inverter working in DCM considering the PV
3308
power utilization and the total harmonic distortion (THD) of
the output current in presented in [15]. Although the
electrolytic capacitors are suitable candidate as power
decoupling storage elements due to its large capacitance and
low cost, it is the major factor limiting the lifetime of the
micro-inverter. Therefore, a three-port flyback converter,
which one port is dedicated to power decoupling [16], and its
advanced version with reduced components [17] have been
proposed to enable a small film capacitor to be used instead of
the bulky and low reliable electrolytic capacitor.
The phase synchronization control strategies of an
interleaved flyback inverter operating in BCM are analyzed in
[18] and in order to reduce the switching loss associated with
the current mismatch in two phases on the inverter, a closed
loop phase synchronization control method is proposed.
According to these efforts to enhance flyback performance, Fig. 1: Circuit configuration of the proposed flyback
this paper proposes a novel adaptive snubber suitable for the micro inverter with the adaptive snubber
flyback converter based on the hybrid operation of BCM and
and also reduce the turn-off switching loss of the main switch.
DCM. The proposed adaptive snubber combines the simplicity
This additional capacitor does not affect the soft-switching of
of RCD snubber and efficiency of active clamp circuits to
the main switch at turn-on if enough time delay is applied
achieve a high efficiency at lowest possible cost.
between zero crossing of magnetizing current and the next
The proposed snubber consists of an auxiliary switch and a
turn-on switching signal.
snubber capacitor, which is similar to the conventional active
Since natural turn on ZVS is not provided during the DCM
clamping circuit, but with different control scheme and design
operation of the flyback micro inverter, the turn-on switching
procedure. The lower snubber capacitance and simple gate
loss associated with the additional capacitor becomes
driving signal for the auxiliary switch make it possible to
dominant and the reduction of turn-off switching loss becomes
implement a low cost and high efficiency flyback micro-
marginal. Therefore, to improve the overall efficiency of the
inverter.
flyback micro-inverter, as shown in Fig. 1, an auxiliary switch
Effectiveness of the proposed snubber is experimentally
validated through experimentation on a 250W flyback micro- is employed to disconnect the snubber capacitor from the
inverter prototype. circuit when the converter works in the DCM mode. As a
result, during the DCM operation of the converter the
auxiliary switch (Sa) is OFF and during the BCM operation,
II. PROPOSED FLYBACK MICRO-INVERTER WITH ADAPTIVE
when the natural turn-on ZVS is achievable, the auxiliary
SNUBBER switch is always ON, effectively adding the capacitor Cs
Fig. 1 shows the proposed flyback micro-inverter with the across the drain-source terminals of the main switch.
novel adaptive snubber which is suitable for use in converter The proposed adaptive snubber provides the same
operating with hybrid conduction. The configuration of the advantages as the conventional active clamping method in
adaptive snubber is similar to the previous active clamping terms of zero voltage switching and also limiting the voltage
circuit. However, the control scheme and the design procedure across the main switch at turn-off at a much lower cost.
of the proposed adaptive snubber are entirely different. Furthermore, the auxiliary switch is operated at only double
The BCM operation of the flyback converter has several line frequency and the required snubber capacitance is very
advantages compared to DCM operation in terms of power small, in range of tens of nano-farad, which is more than ten
density and conversion efficiency. Since the BCM operation times smaller than the required capacitance value in the
provides a natural ZVS turn-on for the main switch, the conventional active clamping. Also, the additional isolated
reduced switching loss allows higher switching frequency and power supply for driving the auxiliary switch is not required
compact design. However, it still has the turn-off switching any more since an N-channel MOSFET can be used that shares
loss that limits the maximum allowable switching frequency. the same source terminal with the main MOSFET. All these
Also, during the BCM operation of the converter and near the advantages reduce the cost and complexity and also increase
peak grid voltage where the instantaneous transferred power is the overall efficiency of the PV micro-inverter system.
high, the leakage inductance in flyback converter causes
voltage overshoot across the main switch at turn-off and the A. Operation Principle
turn-off switching loss becomes more severe. Fig. 2 shows the theoretical operational waveforms of the
proposed flyback micro-inverter under BCM during one
The proposed snubber solves these problems by adding
switching period TS, and Fig. 3 shows the equivalent circuits
additional capacitor in parallel with the main switch. This
for each operational mode.
capacitor reduces the voltage overshoot and also lowers the
To simplify the analysis of the operational modes, only one
rising slope of the turn-off voltage. These effects allow low
flyback converter is considered due to the fact that converters
conduction loss from the usage of lower voltage rating device

3309

TS
S by adding the snubber capacitor Cs and limiting the voltage
Td
Sa
iref
iLm
iD iref /N
vds
Vin+Vo /N
t0 t1 t2 t3 t4 t5 t
Fig. 2: Theoretical operation waveform under BCM.
operate in the same manner in interleaved mode. Also, the
voltages Vin and Vo and the reference current iref are assumed to
be constant during TS. Furthermore, the auxiliary switch Sa is
always in the ON state, since the converter operates under
BCM. Each operation mode is described next.
1) Mode 1 (t0 í t1): Just before t0, the switch S has
turned on, and the magnetizing inductor current iLm increases
linearly from zero until it reaches iref. At this moment, the
switch S turns off and Mode 1 ends. The required time in this
mode is:
Ltot
t 01 = .iref
Vin (1)
Ltot = Lm + Llk ≈ Lm
Although the iref is assumed to be constant during TS, the
actual value of the iref is continuously changing across the grid
period [10], [19], as:
2) Mode 2 (t1 – t2) At t1, switch S turns off and the
resonance between inductors (Lm, Llk) and capacitors (Cs, Coss)
begins where Coss is the parasitic capacitor of the switch S.
The duration of this mode is relatively short compared to the
whole resonant period due to the large value of the
magnetizing inductor current at turn-off, thus the voltage vds
increases almost linearly. This mode ends when the voltage vds
reaches Vin+Vo/N.
3) Mode 3 (t2 – t3): The energy stored in the
magnetizing inductor Lm is transferred to the output side. The
output diode current iD linearly decreases to zero, thus the
reverse-recovery effect of D is minimized. If the leakage
inductance is assumed to be zero as ideal condition to simplify
the operational analysis, the voltage vds maintains the constant
value of Vin+Vo/N during this mode. However, in the real case,
the leakage inductance causes a high frequency oscillation at
the beginning of this mode as it is shown in the dotted
waveforms of Fig. 2. The conduction path of the oscillating
current is also shown in Fig. 3 with dotted lines. Without a
snubber or clamping circuit to protect the main switch, this
3310
voltage spike enforces the use of a high voltage rated
switching device which has normally a high Rds(on). Therefore,
overshoot, a MOSFET with lower voltage rating can be used
for the main switch, resulting in reduction of the conduction
loss.
4) Mode 4 (t3 – t4): This is a resonant mode similar to
Mode 2. If the steady state value vds at during Mode 3 satisfies
the following equation:
Vin + Vo N > 2Vin
(2)
→ Vo > NVin
the voltage vds decreases to zero with a resonant manner
within the maximum of half resonant period regardless of the
amount energy stored in the magnetizing inductance.
5) Mode 5 (t4 – t5): While the current iLm flows
negatively through the anti-parallel diode of the switch S, the
gating signal of the switch S is applied, thus satisfying the
ZVS turn-on. The time delay Td between the zero magnetizing
current and the turn-on of switch S is equal or less than half of
the resonant period.
B. Snubber capacitor selection
The value of snubber capacitor Cs is selected to minimize
the spike voltage across the main switch during Mode 3.
Although a relatively large snubber capacitor can effectively
reduce the spike voltage, it increases the duration of
oscillations during the turn off process which results in low
overall efficiency. Large snubber capacitor also increases the
resonant time of Mode 4 (tQR) which can result in lower
switching frequency and higher output current THD.
Since the snubber capacitor is very small in the order of
nano-Farads, the parasitic capacitance of various circuit
elements can have a great effect on the required snubber
capacitance. As it is described in [9] and [20], the capacitance
of the primary winding of transformer and also the reflected
anode-cathode capacitance of the secondary rectifier are
effectively in parallel with Coss during the switching transients,
so, the most suitable equation for calculating the required
snubber capacitance can be deduced as:
LLk
vspike = iLm _ peak (3)
Coss Cs C pw N 2C D
Where Cpw is the primary winding capacitance of the flyback
transformer and CD is the anode-cathode capacitance of the
secondary rectifier. Based on (3) and the peak inductor current
and also the desired voltage spike, the required snubber
capacitance can be calculated.
III. EXPERIMENTAL RESULTS
To verify the effectiveness of the proposed adaptive
snubber, a 250W flyback micro-inverter prototype has been
built.
Table I and II show the system specifications and the major
component list used in the hardware, respectively. Based on (3)
and the measured values of Cpw=1.88nF, CD=35pF, Coss-
 (a)
(c)
(e)
Fig. 3. Equivalent circuits for each operational mode: (a) Mode 1 [t0-t1]. (b) Mode 2 [t1-t2]. (c) Mode 3 [t2-t3]. (d) Mode 4 [t3-t4]. (e) Mode 5 [t4-t5].
20V=3.8nF (accumulative Coss of two parallel MOSFETs at 20V)
and the maximum overshoot of 130V on the MOSFETs, the
final snubber capacitance was selected as 1.68nF.
The overall control schemes were implemented based on
dsPIC microcontroller dsPIC33FJ16GS504. During the DCM
and BCM operation, the turn-off timings of the main switches
are determined by comparing the reference current with the
instantaneous current of the primary side of each transformer.
The constant frequency operation determines the turn-on
timings of the switches in the DCM. However, in BCM, the
turn-on timing of the main switch S1 is determined by detecting
zero voltage across vds1, instead of sensing the output diode
current, and the turn-on timing of the main switch S2 is
synchronized to the turn-off timing of S1 for simple interleaved
operation.
The experimental results of the micro-inverter with the
proposed adaptive snubber and without considering the MPPT
[21-25] operation are presented in Figs. 4-7 for the output
3311
(b)
(d)
power of 210W which corresponds to the maximum efficiency
of the converter.
Fig. 4 shows the overall operational waveforms across the
grid cycle. The grid current is synchronized with the grid
voltage by Phase-Locked Loop (PLL). The auxiliary switch Sa1
gating signal in Fig. 4 illustrates the 120Hz operation of this
switch and also the DCM-BCM mode transition during the
operation of the converter.
Fig. 5 illustrates the interleaved operation of the converter
in the BCM. It is shown in this figure that the turn-on timing
of the main switch S2 is synchronized with the turn-off timing
of the main switch S1.
Fig. 6 shows the mode transition waveform from DCM to
BCM.
The efficiency of the prototype from the input port to the
output port (including the micro-processor and driving power)
was measured as shown in Fig. 7. The maximum efficiency is
95.8% and the CEC weighted efficiency is 94.6%. The
 TABLE I
S1 (10V/div)
SYSTEM SPECIFICATIONS
Output Power (Po) 250 W Sa1 (10V/div)
Input Voltage (vin) 24 ¥ 35 V vds1 = 60V (50V/div)

Grid Voltage (vg) 240 Vrms

Grid Frequency (fg) 60 Hz
ip1 = 16A (10A/div)
Switching Frequency DCM 100 kHz
(fs) BCM 100 ¥ 300 kHz
t (4us/div)
TABLE II
MAJOR COMPONENT LIST Fig. 6. Gate-to-source voltage of the main switch S1, Gate-to-source
voltage of the auxiliary switch Sa1, drain-to-source voltage of S1 vds1, and
Main Switch (S1, S2) BSC190N15NS3 × 2 primary side current of the transformer ip1 during mode transition
between DCM and BCM.
Output Diode (D1, D2) IDB06S60C
Auxiliary Switch (Sa1, Sa2) IPD320N20N3G
Unfolding Bridge Switch (S3 í S6) TK39J60W5
Core Type RM14 – N97
Turns Ratio (N) 6
Transformer (TR1, TR2) Magnetizing
6.15ȝH, 6.18ȝH
Inductance (Lm1, Lm2)
Leakage Inductance
0.033ȝH, .037ȝH
(Llk1, Llk2)
Snubber Capacitor (Cs) 1.68 nF
Inductor (Lf) 220 ȝH
Output Filter
Capacitor (Cf) 220 nF Fig. 7. Efficiency curve.

efficiency of the converter is also measured using a

vg = 350V (250V/div) conventional RCD snubber (Csnub=1.68nF and Rsnub = 4.3kŸ)
ig = 1.2A (2A/div) and it is shown in Fig. 7. The maximum achieved efficiency in
this case is 94.4% with the CEC efficiency of 92.6%.
S1 (10V/div) IV. CONCLUSION
This work presented a new adaptive snubber for the hybrid
(DCM-BCM) operation of flyback micro-inverter. The
Sa1 (10V/div) proposed adaptive snubber consists of the minimum number
of components, namely, a small auxiliary switch and a small
capacitor. It also exploits the parasitic capacitors of the
t (4ms/div) flyback circuit (the main switch Coss, transformer winding
Fig. 4. Experimental waveforms. Grid voltage vg, grid current ig, drain- capacitor and secondary rectifier capacitor) to further reduce
to-source voltage of S1 vds1, and primary side current of the transformer the required value of the snubber capacitor. Furthermore, the
ip1. auxiliary MOSFET switches at only double-line frequency
and shares the same source terminal with the main switch of
the flyback, which reduces the cost and complexity controller
and the gate drive circuit.
Based on the detailed operation principal of the snubber
circuit and the presented mathematical formulas, the value of
the snubber capacitor can be selected for the best performance.
Also, the consideration for the optimum design aiming to
higher efficiency was presented.
The experimental results on a 250W hardware prototype
demonstrate that the proposed adaptive snubber is able to limit
Fig. 5. Interleaved operation waveforms: Gate-to-source voltage of the
the voltage overshoot across the main switch and achieve high
main switch (S1, S2), drain-to-source voltage of the main switch (vds1, system efficiency through enabling the use of lower voltage
vds2). MOSFET and also providing soft switching. The overall

3312
system with the proposed snubber achieved the maximum interleaved flyback microinverters," IEEE Trans. Power Electron.,
vol.29, no.11, pp. 5921-5932, Nov. 2014.
efficiency of 95.8% and the CEC weighted efficiency of
[19] Q. Mo, M. Chen, Z. Zhang, Y. Zhang, and Z. Qian, “Digitally controlled
94.6% active clamp interleaved flyback converters for improving efficiency in
photovoltaic grid-connected micro-inverter,” in Proc. IEEE 27th Annu.
REFERENCES Power Electron. Conf. Expo., Feb. 2012, pp. 555–562.
[1] Q. Li and P. Wolfs, “A review of the single phase photovoltaic module [20] H. K. Patel, "Voltage transient spikes suppression in flyback converter
integrated converter topologies with three different DC link using dissipative voltage snubbers," in Proc. Of 3rd IEEE Conference
configurations,” IEEE Trans. Power Electron., vol. 23, no. 3, pp. 1320– on Industrial Electronics and Applications (ICIEA) , pp. 897-901, 3-5
1333, May 2008. June 2008.
[2] H. Hu, S. Harb, N. Kutkut, I. Batarseh, and Z. J. Shen, “A review of [21] G. Farivar, B. Ashaei, and S. Mehrnami, “An Analytical Solution for
power decoupling techniques for micro-inverters with three different Tracking Photovoltaic Module MPP,” IEEE J. Photovolt., vol. 3, no. 3,
decoupling capacitor locations in PV systems,” IEEE Trans. Power pp. 1053–1061, Jul. 2013.
Electron., vol. 28, no. 6, pp. 2711–2726, Jun. 2013. [22] G. Falahi, H. Mokhtari, "Performance Improvement of Parallel Active
[3] D. Meneses, F. Blaabjerg, Ó. García, and J. A. Cobos, “Review and Power Filters Using Droop Control Method," Power and Energy
comparison of step-up transformerless topologies for photovoltaic AC- Engineering Conference, 2009. APPEEC 2009. Asia-Pacific , vol., no.,
Module application,” IEEE Trans. Power Electron., vol. 28, no. 6, pp. pp.1,4, 27-31 March 2009.
2649–2663, Jun. 2013. [23] K. Tan, R. Yu, S. Guo, A. Q. Huang, "Optimal design methodology of
[4] M. Gao, M. Chen, C. Zhang, and Z. Qian, "Analysis and implementation bidirectional LLC resonant DC/DC converter for solid state transformer
of an improved flyback inverter for photovoltaic AC module application," Industrial Electronics Society, IECON 2014 - 40th Annual
applications," IEEE Trans. Power Electron, vol.29, no.7, pp. 3428-3444, Conference of the IEEE , vol., no., pp.1657,1664, Oct. 29 2014-Nov. 1
July 2014. 2014
[5] A. C. Kyritsis, E. C. Tatakis, and N. P. Papanikolaou, “Optimum design [24] E. M. Najm, A. Q. Huang, W. Yu, "A novel low cost PV micro-inverters
of the current-source flyback inverter for decentralized grid-connected based on analog control and interleaving method," Applied Power
photovoltaic systems,” IEEE Trans. Energy Convers., vol. 23, no. 1, pp. Electronics Conference and Exposition (APEC), 2015 IEEE , vol., no.,
281–293, Mar. 2008. pp.221,228, 15-19 March 2015
[6] Y.-H. Kim, J.-W Jang, S.-C Shin, and C.-Y Won, "Weighted-efficiency [25] S. Mehrnami, S. Farhangi, "A fast maximum power point tracking
enhancement control for a photovoltaic AC module interleaved flyback technique for PV powered systems," Industrial Electronics, 2009.
inverter using a synchronous rectifier," IEEE Trans. Power Electron., IECON '09. 35th Annual Conference of IEEE , vol., no., pp.42,46, 3-5
vol.29, no.12, pp. 6481-6493, Dec. 2014. Nov. 2009
[7] J. Zhang, X. Huang, X. Wu, and Z. Qian, “A high efficiency flyback
converter with new active clamp technique,” IEEE Trans. Power
Electron., vol. 25, no. 7, p. 1775–1785, Jul. 2010.
[8] A. C. Nanakos, E. C. Tatakis, and N. P. Papanikolaou, “A weighted-
efficiency-oriented design methodology of flyback inverter for AC
photovoltaic modules,” IEEE Trans. Power Electron., vol. 27, no. 7, p.
3221– 3233, Jul. 2012.
[9] Y.-H. Kim, Y.-H. Ji, J.-G. Kim, Y.-C. Jung, and C.-Y. Won, “A new
control strategy for improving weighted efficiency in photovoltaic AC
module-type interleaved flyback inverters,” IEEE Trans. Power
Electron., vol. 28, no. 6, pp. 2688–2699, Jun. 2013.
[10] Z. Zhang, X.-F. He, and Y.-F. Liu, “An optimal control method for
photovoltaic grid-tied-interleaved flyback micro-inverters to achieve
high efficiency in wide load range,” IEEE Trans. Power Electron., vol.
28, no. 11, p. 5074– 5087, Nov. 2013.
[11] N. Sukesh, M. Pahlevaninezhad, and P. K. Jain, "Analysis and
implementation of a single-stage flyback PV microinverter with soft
switching," IEEE Trans. Ind. Electron, vol.61, no.4, pp.1819,1833,
April 2014.
[12] Y. Li and R. Oruganti, “A low cost flyback CCM inverter for AC
module application,” IEEE Trans. Power Electron., vol. 27, no. 3, pp.
1295–1303, Mar. 2012.
[13] F. Fongang Edwin, W. Xiao, and V. Khadkikar, “Dynamic modeling
and control of interleaved flyback module-integrated converter for PV
power applications,” IEEE Trans. Ind. Electron., vol. 61, no. 3, pp.
1377–1388, Mar. 2014.
[14] T. V. Thang, N. M. Thao, J.-H Jang, and J.-H Park, "Analysis and
design of grid-connected photovoltaic systems with multiple-integrated
converters and a pseudo-DC-link inverter," IEEE Trans. Ind. Electron.,
vol.61, no.7, pp. 3377-3386, July 2014.
[15] S. Zengin, F. Deveci, and M. Boztepe, “Decoupling capacitor selection
in DCM flyback PV microinverters considering harmonic distortion,”
IEEE Trans. Power Electron., vol. 28, no. 2, pp. 816–825, Feb. 2013.
[16] H. Hu, S. Harb, X. Fang, D. Zhang, Q. Zhang, Z. J. Shen, and I.
Batarseh, “A three-port flyback for PV micro-inverter applications with
power pulsation decoupling capability,” IEEE Trans. Power Electron.,
vol. 27, no. 9, pp. 3953–3964, Sep. 2012.
[17] H. Hu, S. Harb, N. H. Kutkut, Z. J. Shen, and I. Batarseh, “A single-
stage micro-inverter without using electrolytic capacitors,” IEEE Trans.
Power Electron., vol. 28, no. 6, pp. 2677–2687, Jun. 2013.
[18] Zhang, M. Chen, W. Chen, C. Jiang, and Z. Qian, "Analysis and
implementation of phase synchronization control strategies for BCM

3313
 energies
Article
A Phase-Shift-Modulated LLC-Resonant
Micro-Inverter Based on Fixed
Frequency Predictive-MPPT
Omar Abdel-Rahim 1, * , Nehmedo Alamir 1 , Mohamed Abdelrahem 2,3, * , Mohamed Orabi 1 ,
Ralph Kennel 2 and Mohamed A. Ismeil 2,4
1 Aswan Power Electronics Applications Research Center (APEARC), Faculty of Engineering, Aswan
University, Aswan 81542, Egypt; nehmedo.alamir@aswu.edu.eg (N.A.); morabi@apearc.aswu.edu.eg (M.O.)
2 Institute for Electrical Drive Systems and Power Electronics (EAL), Technical University of Munich (TUM),
80333 Munich, Germany; ralph.kennel@tum.de (R.K.); melzanaty@apearc.aswu.edu.eg (M.A.I.)
3 Electrical Engineering Department, Faculty of Engineering, Assiut University, Assiut 71516, Egypt
4 Electrical Engineering Departement, Faculty of Engineering, South Valley University, Qena 83523, Egypt
* Correspondence: o.abdelrahim@aswu.edu.eg (O.A.-R.); mohamed.abdelrahem@tum.de (M.A.)




Received: 17 February 2020; Accepted: 17 March 2020; Published: 20 March 2020


Abstract: Maximum Power Point Tracking (MPPT) control is an essential part of every photovoltaic
(PV) system, in order to overcome any change in ambient environmental conditions and ensure
operation at maximum power.. Recently, micro-inverters have gained a lot of attention due to their
ability to track the true MPP for each individual PV module, which is considered a powerful solution
to overcome the partial shading and power mismatch problems which exist in series-connected
panels. Although the LLC resonant converter has high efficiency and high boosting ability, traditional
MPPT techniques based on Pulse Width Modulation (PWM) do not work well with it. In this paper,
a fixed frequency predictive MPPT technique is presented for the LLC resonant converter to be used
as the first-stage in a PV micro-inverter. Using predictive control enhances the tracking efficiency and
reduces the steady state oscillation. Operation with fixed switching frequency for the LLC resonant
converter improves the total harmonic distortion profile of the system and ease the selection of circuit
magnetic component. To demonstrate the effectiveness of the proposed MPPT technique, the system is
simulated using MATLAB/Simulink platform. Furthermore, a 150 W hardware prototype is developed
and tested. Both simulation and experimental results are consistent and validate the proper operation
of the developed system.

Keywords: micro-inverter; LLC resonant converter; MPPT; phase-shift control; fixed frequency

1. Introduction
Photovoltaic (PV) cells are one of the promising kinds of renewable energy. The compound annual
growth rate (CAGR) of PV installations was 36.8% between the years 2010 to 2018 [1,2]. It is worth
noting that the annual installation of PV stations in 2017 alone was the same as the total PV capacity
until the end of 2012 [3]. It is expected that renewable power capacity will continue to increase by
50% between 2019 and 2024 [4]. PV systems have different classifications, but they could be classified
according to grid integration into [5–7]:

(1) Stand-alone PV systems (off-grid), and

(2) Grid-connected PV systems.

In the stand-alone (off-grid) mode, the PV system is isolated from the main grid, and hence the
PV-voltage is converted to AC-voltage to feed AC-loads. Battery storage is an essential component

Energies 2020, 13, 1460; doi:10.3390/en13061460 www.mdpi.com/journal/energies

Energies 2020, 13, 1460 2 of 16

in off-grid PV systems, in order to feed essential loads during emergencies [8]. There are different
categories of storage systems like lead acid batteries. For example, the usage of supercapacitors is a
good solution in case of limited energy storage capacity [9]. On the other hand, when the storage
station is not able to cover all load demands, a hybrid system can represent a good solution [10].
Hybrid system using PV generation and hydrogen energy storage system is presented in [11].
In grid integrated PV systems, the PV panels are connected to the grid via a DC/AC converter.
The grid voltage is fixed, and hence, the PV system is working as a current source which is injecting a
sinusoidal current into the grid. The different formations which are developed in the literature for
grid-tied PV systems are summarized as follows:
(i) Centralized PV arrays, where a huge number of PV panels are connected in series and parallel
configuration and then they are connected to a centralized DC/AC converter. This configuration is
commonly used in very high power PV plants. However, this system requires expensive switches and
high voltage wiring [12], and in case of partial shading or mismatches, the system fails to extract the
maximum power [12,13].
(ii) Multi-string inverter. This configuration is used in medium power PV power plants. Strings
are connected in parallel and each group of the parallel string is connected to their own DC-DC
converter. All these converters are connected to the utility grid through a central DC/AC converter.
Partial shading problems have less effect than in the centralized formation case, however as it still
depends on a centralized inverter, inverter failures can shut down the whole plant [13].
(iii) Micro-inverter. This configuration is well suited for low power PV applications, where each
PV panel has its own DC/AC converter [14–16]. Such a formation overcomes partial shading and
power mismatch problems. Due to their compactness and high efficiency, micro-inverters have gained
a lot of attention in grid-tie applications [17–19]. The main advantage of using a micro-inverter is that
each PV module has its own power conversion, and hence independent MPPT control. Consequently,
a global MPPT system is not required, as each PV module is operating at its maximum power in an
independent way. Micro-inverters can be classified into single-stage micro-inverters and two-stage
microinverters [20]. Two stage micro-inverters consist of DC-DC converters to perform voltage boosting
and MPPT extraction and DC-AC converter to convert the DC voltage into AC voltage, while single
stage micro-inverters consist of only one stage that performs voltage boosting, MPPT tracking and
converts the DC voltage into AC voltage.
The general structure of a micro-inverter is illustrated in Figure 1. The general requirements of
micro-inverters include the ability to operate at high efficiency, high power density and operation over
a wide range of dynamic load changes [21]. The first stage of a micro-inverter is constituted by a DC-DC
converter. Different types of DC-DC converters are described in the literature, but the high frequency
(HF) resonant converter has small-size magnetic components, galvanic isolation, and higher efficiency
over a wide load range. Such features make resonant converters a perfect choice for micro-inverter
applications. Soft switching techniques are applied to the resonant converter to improve the overall
system efficiency and reduce the switching losses of the system. Micro-inverter topologies can be
classified into isolated micro-inverters and non-isolated micro-inverters. In non-isolated micro-inverters,
the configuration does not include any transformer and the DC/DC stage is used such as boost,
buck-boost or cuk converters [22–24]. Isolated micro-inverters include a transformer, high frequency
or line frequency transformer, such as current-feed push-pull [25,26], flyback converters [27,28], matrix
converters [29,30] and resonant converters [31–34]. Converters operating in soft switching mode are
preferred more than hard-switching converters. Hard-switching causes higher switching losses, lower
efficiency and damage to the switching devices. Resonant converters represent a superior solution
for PV applications due to their high efficiency and zero-voltage switching (ZVS) and/or zero-current
switching (ZCS) operation [35,36].
Different MPPT techniques for PV applications have been reported in the literature like perturb
and observe (P&O), incremental conductance (INC), or fuzzy logic MPPT. MPPT techniques based
on predictive control have been introduced in [37–40]. Fixed-frequency fixed-step INC predictive
Energies 2020, 13, 1460 3 of 16

MPPT proposed in [37] with an enhancement of the ordinary INC method by predicting the error
Energies 2020, 13, x FOR PEER REVIEW 3 of 15
before applying the switching signal. In [38], a variable-frequency fixed-step P&O predictive control
method was implemented with a good dynamic response and small power ripple at steady state,
while a variable step size sensorless predictive MPPT was proposed in 39]. In [40], a predictive MPPT
while a variable step size sensorless predictive MPPT was proposed in [39]. In [40], a predictive
is presented for a Z-source grid-connected inverter using variable frequency control. All the
MPPT is presented for a Z-source grid-connected inverter using variable frequency control. All the
previously mentioned techniques are implemented for PWM converters, but none of them could be
previously mentioned techniques are implemented for PWM converters, but none of them could be
applied directly to resonant topologies. Due to the nonlinearity of the voltage gain versus frequency
applied directly to resonant topologies. Due to the nonlinearity of the voltage gain versus frequency
relationship, MPPT with a special design mast be considered for frequency-modulated LLC resonant
relationship, MPPT with a special design mast be considered for frequency-modulated LLC resonant
converters. Most of the old MPPT methods have been modified to be compatibile with different
converters. Most of the old MPPT methods have been modified to be compatibile with different
resonant converters. A modified P&O for a resonant converter is presented in [41,42]. Although this
resonant converters. A modified P&O for a resonant converter is presented in [41,42]. Although this
method is easy, it has some limitations that can be summarized as: fixed-step P&O delays the
method is easy, it has some limitations that can be summarized as: fixed-step P&O delays the response
response of the system, and the long tracking time leads to an increase in power losses.
of the system, and the long tracking time leads to an increase in power losses.
In [43] a fixed step P&O variable frequency control method is proposed for the LLC resonant
In [43] a fixed step P&O variable frequency control method is proposed for the LLC resonant
converter. Although it is a simple technique, it suffers from a long transient period. A variable step
converter. Although it is a simple technique, it suffers from a long transient period. A variable step
variable frequency MPPT control is implemented in [44], but it suffers from a wide range of
variable frequency MPPT control is implemented in [44], but it suffers from a wide range of harmonics
harmonics and bigger EMI, due to its variable frequency operation. Also, the design of the magnetic
and bigger EMI, due to its variable frequency operation. Also, the design of the magnetic component is
component is a cumbersome operation; a phase-shift fixed-frequency modulation control is used to
a cumbersome operation; a phase-shift fixed-frequency modulation control is used to overcome those
overcome those disadvantages [44,45]. The use of fixed-frequency modulation with the LLC resonant
disadvantages [44,45]. The use of fixed-frequency modulation with the LLC resonant converter is first
converter is first addressed in [46] and 47], which use a fixed-frequency P&O MPPT but with high
addressed in [46,47], which use a fixed-frequency P&O MPPT but with high steady state oscillation.
steady state oscillation.

Figure 1. A two-stage PV grid-connected system.

This manuscript presents

This manuscript presentsa anovel novel fixed-frequency
fixed-frequency phase-shift
phase-shift (FFPS)
(FFPS) predictive
predictive MPPT MPPT for
for LLC
LLC resonant
resonant converters
converters to betousedbe usedin thein two-stage
the two-stage micro-inverter
micro-inverter configuration
configuration for grid-tied
for grid-tied PV
PV applications. This manuscript modifies the content developed in [48],
applications. This manuscript modifies the content developed in 48], where the system is briefly where the system is briefly
discussed
discussed andandonlyonlysimple
simplesimulation
simulation results are included.
results On theOn
are included. other
thehand,
otherhere complete
hand, here hardware
complete
results
hardwareare results
presented, as well as an
are presented, as extensive
well as ansystem
extensiveanalysis.
system analysis.
The
The proposed
proposedfixed frequency
fixed frequency predictive MPPTMPPT
predictive for LLCfor
resonant convertersconverters
LLC resonant has severalhasadvantages
several
over other MPPT
advantages techniques
over other MPPTsuitable to besuitable
techniques applied toforbetLLC resonant
applied converters,
for tLLC resonantwhich are as follows:
converters, which
(1) small steady state error due to the use of a PI controller. (2) Fast transient
are as follows: 1) small steady state error due to the use of a PI controller. 2) Fast transient performance due to the
prediction of reference voltage, and (3) simple parameter design due to the
performance due to the prediction of reference voltage, and 3) simple parameter design due to the usage of fixed frequency
modulation.The developed
usage of fixed frequency two-stage micro-inverter
modulation.The developed schematic
two-stage is shown in Figure
micro-inverter 2. It is constructed
schematic is shown in
from a LLC resonant converter and an h-bridge inverter. This manuscript considers
Figure 2. It is constructed from a LLC resonant converter and an h-bridge inverter. This manuscript only the DC-DC
part. The choice
considers only theof the LLC resonant
DC-DC part. Theconverter,
choice ofthe thefirst
LLCstage in the proposed
resonant converter,system, comes
the first stagefrom its
in the
distinct
proposed features
system,overcomes other resonant
from converter
its distinct topologies
features suchresonant
over other as the ability to operate
converter at ZVS/ZCS,
topologies such as
wide input-voltage
the ability to operate range, step-up wide
at ZVS/ZCS, ability, and high efficiency.
input-voltage range, step-up ability, and high efficiency.
This paper is organized
organized as follows:
follows: Section 2 discusses the analysis of LLC resonant DC/DC
converter,
converter,thenthenSection
Section 3 discusses
3 discusses the proposed
the proposed MPPTMPPT technique using the
technique predictive
using Fixed-Frequency
the predictive Fixed-
phase-shift control, Section 4 presents the simulation results for the
Frequency phase-shift control, Section 4 presents the simulation results for the proposed proposed MPPT technique and
MPPT
experimental
technique andresults, and finally
experimental results,Section 5 presents
and finally Sectionthe experimental
5 presents validation of
the experimental the proposed
validation of the
MPPT technique.
proposed MPPT technique.

2. Analysis of the LLC Phase-Shift Modulated Resonant Converter

The LCC resonant converter depicted in Figure 2 consists of s phase-shift switching network,
resonant tank, high frequency transformer, and voltage doubler. The switching network depends on
the voltage gain requirements, and it may be constructed from a half-bridge or full-bridge. The
switching network chops the input DC voltage into high-frequency square pulses. The resonant
network consists of a series of Ls and Cs , and Cp in parallel with the rectifier input. The energy applied
Energies 2020, 13, 1460 4 of 16

2. Analysis of the LLC Phase-Shift Modulated Resonant Converter

The LCC resonant converter depicted in Figure 2 consists of s phase-shift switching network,
resonant tank, high frequency transformer, and voltage doubler. The switching network depends on the
voltage gain requirements, and it may be constructed from a half-bridge or full-bridge. The switching
network chops the input DC voltage into high-frequency square pulses. The resonant network consists
of a series of Ls and Cs , and Cp in parallel with the rectifier input. The energy applied across the
resonant tank by the switching network is circulated through the resonant tank network. Applied
energy is delivered to the output either in the same or the next switching cycle.
The gating signal and key waveforms of the LLC resonant converter are shown in Figure 3.
The converter is analyzed based on the fundamental harmonic approximation (FHA) technique given
in [44]. The equivalent phasor circuit of the LLC converter is illustrated in Figure 4. The DC-input
voltage is converted to a square wave waveform by the switching network. The output of the switched
network appears between X and Y terminals. The root mean square (RMS) value of the fundamental
component of the AC voltage VXY is given by:
√
2
VXY = Vpv (1 − cosδ) (1)
π
where δ is the pulse width angle. it has an angle complementary to the phase shift angle α and
α = π − δ as shown in Figure 2. By using high switching transformer with turns ration n, the RMS
value of primary voltage VLp can be obtained as:

1
VLp = nVo (ac) (2)
2
Then from (1) and (2):
VLp  
= (2/(1 − cos δ)) nVo (ac)/Vpv (3)
VXY
Referring to the phasor equivalent circuit of the converter in Figure 4, the following relation can
be obtained:
VLp 1
= (4)
VXY1 o2 n 2  o  1
π 1 2 2
n
Ls 2
1 + ( LP )(1 − F ) + 8 Q F − F

Thus from (3) and (4), the converter voltage gain M can be obtained by (5) as follows:

n(1 − cosδ)/2
M=  (5)
o2 n 2  o2  12
1 + ( LLPs )(1 − F2 ) + π8 Q F − F1
n

ωs f R
where: Q = ωr RLs0 , F = = fsr , ωs = 2π fs , R0l = n2l .
ωr
l √
Here ωr = 2π fr = 1/ Ls Cs , and Rac = ( π82 )R0l . fs refers to the switching frequency, and fr : is the
resonant frequency. Vo (ac) is the RMS of the output voltage. Vxy1 is the r.m.s. value of the output
voltage of the inverter across terminals XY. R0l is the load resistance referred to primary-side.
 Applied
Appliedenergy
energy is is delivered
delivered to
to the
the output either in
output either in the
the same
same ororthe
thenext
nextswitching
switchingcycle.
cycle.
The
The gating signal and key waveforms of the LLC resonant converter are shown ininFigure
gating signal and key waveforms of the LLC resonant converter are shown Figure3.3.The
The
converter
converter is analyzed based on the fundamental harmonic approximation (FHA) technique given inin
is analyzed based on the fundamental harmonic approximation (FHA) technique given
44].
44].The
Theequivalent
equivalentphasorphasor circuit
circuit of
of the LLC converter
the LLC converter is isillustrated
illustratedin
inFigure
Figure4.4.The
TheDC-input
DC-inputvoltage
voltage
isis converted
converted to
to aa square
square wave
wave waveform
waveform by the
by the switching
switching network.
network. The
The output
outputofofthetheswitched
switched
Energies 2020, 13, 1460 5 of 16
network
networkappears
appears between
between XX and
and Y Y terminals. The root
terminals. The root mean
meansquare
square(RMS)
(RMS)value
valueofofthe
thefundamental
fundamental
component
componentof of the
the ACAC voltage
voltage VXYXY is given by:

Phase
Phase shift
shift full Resonant
Resonant HF
HF
bridge
bridge converter
converter Tank
Tank Transformer
Trans former
}} S3
SS11
Ls Css
C 1: nn
1:
Y
Y DC-AC
DC-AC
ILs
Ls
LLpp Converter
Converter Grid
Grid
X
SS44 S2

SS11
IIPV
PV VPV
V (k+1)
PV(k+1) α SS22 Vol
Voltage
tage
α Modified
Modified Gate
Gate Doubler
V MPC-MPPT
MPC-MPPT PI
PI SS33 Doubler
VPV PV +
+ --
VPV signal
signal circuit
circuit SS44 Circuit
Circuit
V PV

Figure 2. Proposed two-stage micro-inverter with LLC resonant converter.

Figure 2.
Figure 2. Proposed
Proposed two-stage
two-stage micro-inverter
micro-inverter with
withLLC
LLCresonant
resonantconverter.
converter.

Energies 2020, 13, x FOR PEER REVIEW 5 of 15

= ( ) (2)

Then from (1) and (2):

Figure 3. Switches gate signals voltage and current waveforms for the LLC resonant converter at
Figure
Figure 3. phase-shift.
Switches
3. Switches gatesignals
gate signalsvoltage
voltage and
and current
current waveforms
waveformsforforthe
theLLC
LLCresonant
resonantconverter at at
converter
arbitrary
= (2 ⁄ (1 − cos ))( ( ) ⁄ ) (3)
arbitrary phase-shift.
arbitrary phase-shift.
√2 (1)
=jX√2LS jX(1
CS− )
= (1 − ) (1)
X
where δ is the pulse width angle. it has an angle complementary to the phase shift angle α and α=π-
where δ is the
δ as shown inpulse
Figurewidth
2. Byangle.
using it
high
ILSswitching
has an angle complementary
transformer withtoturns
the phase
rationshift angle
n, the RMSα value
and α=π-
of
δprimary
as shown in Figure can
voltage 2. Bybeusing highas:switching transformer
obtained Rac with turns
VLP ration n, the RMS value of
jXLP
primary voltage can be obtained as:

4. Phasor equivalent circuit for LLC resonant converter by using FHA.

FigureFigure
4. Phasor equivalent circuit for LLC resonant converter by using FHA.

Referring to the phasor equivalent circuit of the converter in Figure 4, the following relation can
be obtained:

= 1 (4)
1
1+( )(1 − ) + ( − )
8
Thus from (3) and (4), the converter voltage gain M can be obtained by (5) as follows:
( )/
 Energies 2020, 13, 1460 6 of 16

3. Proposed Fixed-Frequency Phase-Shift Modulated MPC-MPPT Control

Unlike the previously mentioned works, this manuscript presents a novel fixed frequency
predictive maximum power point technique with an adaptive step-size for a phase-shift modulated
LLC resonant converter. The schematic of the control feedback system is depicted in Figure 5.
The operation of the proposed methodology is performed in four steps, as follows:
Step 1: The PV terminals voltage value VPV (K), may be increased or decreased, depending on the
applied switching state, during the next sampling time (K+1), And hence the next sampling time of PV
voltage VPV (K+1) is defined as:

VPV (K + 1)1 = VPV (K) + ∆V

(
(6)
VPV (K + 1)2 = VPV (K) − ∆V

where ∆V is the step size. This step size could be set as a fixed value or could be designed to be variable
according to change in the power level. A variable step size could be defined from the following
equation:  
∆V = σPmax (K + 1) − PPV (K ) (7)

where σ is normalizing weigh factor, which is tuned to get good steady state performance by trial and
error, Pmax (K + 1) is the next sampling time predicted maximum extracted PV power, the steps for
calculating the value of Pmax (K + 1) will be explained in this section later. However, at starting an
Energies
initial2020,
value13, x ∆VPEER
ofFOR REVIEW
can be set. 6 of 15

Figure 5. 5.
Figure Block diagram
Block of of
diagram thethe
proposed fixed
proposed frequency
fixed phase-shift
frequency MPPT
phase-shift predictive
MPPT control.
predictive control.

Step2:2:The
Step Thenext
nextsampling
samplingtimetimeofofthe
thePVPVterminals voltage VPV
terminalsvoltage ( (+
K+ 1)1)and
andthe
thePVPVcurrent
current
IPV( (K + 1 )
+ 1) are predicted. In order to reduce the number of sensors, a digital observer is used. Figure. 6
are predicted. In order to reduce the number of sensors, a digital observer is used. Figure
6 presents the equivalent
equivalent circuit
circuit of
of PV
PVmodule
moduleusing
usingthetheThevenin
Thevenintheorem.
theorem.Designed
Designeddigital
digitalobserver
observer
isispresented
presented to to calculate
calculate thethe required
required equivalent
equivalent voltage
voltage ( (Veq)
) andand the equivalent
the equivalent resistance
resistance ( )(Req)
as
as follows:
foellows:
VPV (k) − VPV (k − 1)
Req = − (8)
V ( IPV) −(kV) −(IPV−(k1)− 1)
= − (8)
V I = (V ) −(kI ) +( R− I1)
eq PV eq PV (K) (9)
= (k) + ( ) (9)

Then the next sampling time predicted PV current can be calculated as:
( )− ( + 1) (10)
⎧ ( + 1) =
⎪ ( )
⎨ ( )− ( + 1)
⎪ ( + 1) =
⎩ ( )
Step 3: Next the predicted next sampling time PV that can be extracted from the PV is calculated
as follows:
is presented to calculate the required equivalent voltage ( ) and the equivalent resistance ( ) as
foellows:

V ( ) − V ( − 1)
= − (8)
I ( ) − I ( − 1)
Energies 2020, 13, 1460 7 of 16
= (k) + ( ) (9)
Thenthe
Then thenext
nextsampling
samplingtime
timepredicted
predicted PV
PV current
current can
can be
be calculated
calculated as:
as:
( ) V−eq (K)−V
( PV+(K1)
+1)1 (10)
⎧

( +
IPV1)(K =
+ 1)1 =
⎪ ( R) (K)

 eq
(10)

 eqV (K)−V (K +1)
IPV (K + 1)2( =) − R( (PVK+) 1) 2

⎨


⎪ ( + 1) = eq
⎩ ( )
Step 3: Next the predicted next sampling time PV that can be extracted from the PV is calculated
Step 3: Next the predicted next sampling time PV that can be extracted from the PV is calculated
as follows:
as follows: (
PPV (K + 1)1 = IPV (K + 1)1 × VPV (K + 1)1
(11)
(PPV+(K
1)+= 1)2 =( IPV
+(1)
K +×1)2 ×(VPV + (1)
K + 1)2 (11)
( + 1) = ( + 1) × ( + 1)

Veq

Figure
Figure6.6.The
TheThevenin
Theveninequivalent
equivalentcircuit
circuitof
ofthe
thePV
PVmodule.
module.

After calculating
After calculating all
all possible
possibleoutput
outputpower
powerofof
each switching
each state,
switching during
state, the the
during nextnext
sampling time.
sampling
Maximum
time. Maximumpower couldcould
power be calculated as: as:
be calculated
( + 1) = ( ( + 1) , ( + 1) (12)
Pmax (K + 1) = MAX(PPV (K + 1)1 , PPV (K + 1)2 (12)
Step 4: Then the cost function can be evaluated as:
Step 4: Then the cost function can be evaluated as:

J[1,2] = PPV (K + 1)[1,2] − PPV (K) (13)

After evaluation of the cost function by the proposed FFPS predictive MPPT to track the maximum
available power from the PV, a reference voltage will be chosen for the next sampling period, that the
larger value of cost function will be the next sampling time target, as example, if the cost function J1
is greater than J2 , the PV voltage must be shifted to VPV (k + 1)1 , else if J2 is greater than J1 so the PV
reference voltage will be VPV (k + 1)2 . PI controller is added to track the measured PV voltage with the
predicted reference voltage. Prediction of the reference voltage improves the transient performance
of the system, while using PI controller to track the reference voltage improves the steady state
performance of the system.

4. Simulation Results
The operation of the proposed predictive MPPT technique is validated using a 300 W MATLAB/
Simulink model. The input voltage was changed over a wide range from 25 V to 41 V. The resonant
frequency Fr is selected to be 140 kHz, while switching frequency is set to 154 kHz. The resonant
parameters are set to 4.1/3415 µH/nF for the resonant inductor and resonant capacitor, respectively.
The transformer turns ratio is 1:7. The PV module used in the simulation has maximum power
characteristics (9.6 A/31.25 V). In order to demonstrate the superior behavior of the proposed technique,
different operating condition are taken into account. Figure 7 illustrates a case of study of using fixed
step-size, the irradiation level is set to 100% at the start, then, at time t = 0.7 s, a reduction of the
radiation level to 50% is applied as shown in Figure 7a. The proposed technique can extract maximum
power. Figure 7b,c shows the change of the phase-shift to track the maximum power.
parameters are set to 4.1/3415 μH/nF for the resonant inductor and resonant capacitor, respectively.
The transformer turns ratio is 1:7. The PV module used in the simulation has maximum power
characteristics (9.6 A/31.25 V). In order to demonstrate the superior behavior of the proposed
technique, different operating condition are taken into account. Figure 7 illustrates a case of study of
using fixed step-size, the irradiation level is set to 100% at the start, then, at time t = 0.7 s, a reduction
Energies 2020, 13, 1460 8 of 16
of the radiation level to 50% is applied as shown in Figure 7a. The proposed technique can extract
maximum power. Figure 7b,c shows the change of the phase-shift to track the maximum power.

Figure 7. Proposed fixed-step FFPS predictive MPPT control: (a)the irradiation profile (b) the PV input
Figure 7. Proposed fixed-step FFPS predictive MPPT control: (a)the irradiation profile (b) the PV input
power and the maximum power (c) the power oscillation (d) the phase-shift change.
power and the maximum power (c) the power oscillation (d) the phase-shift change.

Another case study is taken into account where a variable step-size FFPS predictive MPPT is
considered. The results are depicted in Figure 8. The solar irradiation is shown in Figure 8a, the power
extracted from the PV panel is shown in Figure 8b, as it is indicated in Figure 8c the oscillation around
the MPPT is reduced with oscillations of less than ±1%.
Another simulation case study is depicted in Figure 9, where the solar irradiation was changed
from 100% to 30% and then back again to 100% (Figure 9a). As can be seen from Figure 9b, the FFPS
predictive MPPT technique is able to extract the maximum power at all different conditions. The phase
shift variation for this case study is depicted in Figure 9c.
 Energies 2020, 13, x FOR PEER REVIEW 8 of 15

Another case study is taken into account where a variable step-size FFPS predictive MPPT is
considered. The results are depicted in Figure 8. The solar irradiation is shown in Figure 8a, the power
Energies 2020, 13, 1460 9 of 16
extracted from the PV panel is shown in Figure 8b, as it is indicated in Figure 8c the oscillation around
the MPPT is reduced with oscillations of less than ±1%.

Figure 8. The
Figure proposed
8. The variable
proposed variablestep
stepFFPS
FFPS predictive MPPTcontrol:
predictive MPPT control:
(a) (a)
thethe irradiation
irradiation profile
profile (b) the(b) the
Energies 2020,
PV input 13, x
power FOR PEER
and theREVIEW
maximum power (c) the power oscillation (d) the phase-shift
PV input power and the maximum power (c) the power oscillation (d) the phase-shift change.change. 9 of 15

Another simulation case study is depicted in Figure 9, where the solar irradiation was changed
from 100% to 30% and then back again to 100% (Figure 9a). As can be seen from Figure 9b, the FFPS
predictive MPPT technique is able to extract the maximum power at all different conditions. The
phase shift variation for this case study is depicted in Figure 9c.

Figure
Figure 9. 9. Irradiationstep
Irradiation stepvariation
variation(100%-30%-100%)
(100%-30%-100%)with
withproposed
proposedFFPS
FFPSpredictive
predictiveMPPT
MPPTcontrol:
control:
(a) the irradiation profile (b) the PV input power and the maximum power (c) the phase-shift
(a) the irradiation profile (b) the PV input power and the maximum power (c) the phase-shift change. change.

5. Experimental Results
A laboratory prototype for the LLC resonant converter with its proposed FFPS Predictive MPPT
technique was implemented. A general schematic for the hardware layout is depicted in Figure 10. It
consists of a LLC resonant converter, TMS320F2835 DSP controller kit, and resistive load, in addition
 5. Experimental Results
A laboratory prototype for the LLC resonant converter with its proposed FFPS Predictive MPPT
technique was implemented. A general schematic for the hardware layout is depicted in Figure 10. It
consists of a LLC resonant converter, TMS320F2835 DSP controller kit, and resistive load, in addition
Energies
to2020, 13, terminals.
the PV 1460 A JWP 250 PV module is used and the key parameters of the prototype hardware 10 of 16
are listed in Table 1.

5. Experimental Results Table 1. Experimental prototype parameters.

A laboratory prototype for the Rated

LLC Power
resonant converter with
250 Wits proposed FFPS Predictive MPPT
technique was implemented. ATransformer Core Size for the hardware
general schematic RM14 layout is depicted in Figure 10.
Turns Ratio 7
It consists of a LLC resonant converter, TMS320F2835 DSP controller kit, and resistive load, in addition
to the PV terminals. A JWP 250Magnetizing
PV moduleInductance 20.5 μH
is used and the key parameters of the prototype hardware
Switch S1-S4 IRFB4510PBF
are listed in Table 1.
Output Rectifier D1-D2 QH03TZ600

Figure 10. The experimental setup for the LLC resonant converter.
Figure 10. The experimental setup for the LLC resonant converter.
Table 1. Experimental prototype parameters.

Rated Power 250 W

Transformer Core Size RM14
Turns Ratio 7
Magnetizing Inductance 20.5 µH
Switch S1-S4 IRFB4510PBF
Output Rectifier D1-D2 QH03TZ600

Energies 2020, 13, x FOR PEER REVIEW 10 of 15

Different study cases have been investigated in the laboratory for the proposed FFPS predictive
MPPT forDifferent
the LLCstudy cases converter,
resonant have been investigated
and one ofinthese
the laboratory for the proposed
cases is depicted FFPS
in Figure 11.predictive
In this study
MPPT for the LLC resonant converter, and one of these cases is depicted
case, the available maximum power from the real system is 100 W. As can be seen from in Figure 11. In this study
the graph,
case, the available maximum power from the real system is 100 W. As can be seen from the graph,
the proposed algorithm is efficiently able to extract the 100 W with efficiency around 99%. The steady
the proposed algorithm is efficiently able to extract the 100 W with efficiency around 99%. The steady
state oscillation for the proposed phase-shift MPPT technique is illustrated in Figure 12, where it is
state oscillation for the proposed phase-shift MPPT technique is illustrated in Figure 12, where it is
shown the power
shown oscillates
the power around
oscillates the
around theMPP
MPPwithin ±0.5WWrange.
within aa ±0.5 range.

Figure 11. 11.PV

Figure PVvoltage current
voltage current andand power
power for thefor the proposed
proposed fixed-frequency
fixed-frequency predictive
predictive MPPT MPPT
(Pmax=100
(PmaxW)= (Ch2
100 W) (Ch2
(green) (green)
is PV is Ch4
voltage, PV voltage,
(pink line)Ch4 (pink
is the line) isthe
PV current), thePV
PV current),
extracted the (lavender
power PV extracted
powerline).
(lavender line).
 Figure 11. PV voltage current and power for the proposed fixed-frequency predictive MPPT (Pmax=100
W) (Ch2 (green) is PV voltage, Ch4 (pink line) is the PV current), the PV extracted power (lavender
Energies 2020, 13, 1460 11 of 16
line).

Steady state
Figure 12. Steady state PV
PVpower
poweroscillations
oscillationsfor
forthe
theproposed
proposed fixed-frequency
fixed-frequency predictive
predictive MPPT (Ch2
(green) is PV voltage, Ch4 (pink line) is the PV current), the
the PV
PV extracted
extracted power
power (lavender
(lavender line).
line).

Another experimental
Another experimental casecase study
study isis depicted
depicted in
in Figure
Figure 13. In this
13. In this study
study case,
case, the
the available
available
maximum power from the real system is 115 W. As can be seen from the graph,
maximum power from the real system is 115 W. As can be seen from the graph, the proposed the proposed algorithm
is efficiently
algorithm
Energies able
2020,is13, to PEER
extract
efficiently
x FOR ablethe 115 W effectively.
to extract
REVIEW the 115 W effectively. 11 of 15

Figure 13.
Figure 13. PV
PV voltage
voltage current
current and
and power
power for the proposed
for the proposed Fixed-Frequency
Fixed-Frequency predictive
predictive MPPT
MPPT
(Pmax=115 W) (Ch2 (green) is PV voltage, Ch4 (pink line) is the PV current), the PV extracted power
(P max = 115 W) (Ch2 (green) is PV voltage, Ch4 (pink line) is the PV current), the PV extracted power
(lavender line).
(lavender line).

To prove
To prove the
the validity
validity ofof the
the proposed
proposed MPPTMPPT technique
technique under
under partial
partial shading
shading conditions
conditions andand
abrupt change
abrupt changeininsolar
solar radiation
radiation conditions,
conditions, a partial
a partial shading
shading test
test and and change
abrupt abrupt inchange in solar
solar radiation
radiation
test test are performed,
are performed, illustrated illustrated
in Figures 14 in and
Figure
15. 14
In and Figure shading
the partial 15. In thecasepartial shading
of study, case in
depicted of
study, depicted in Figure 14, a part of the real PV module is under lower radiation
Figure 14, a part of the real PV module is under lower radiation comditions. The radiation level and the comditions. The
radiation level
maximum and the
monitored maximum
available powermonitored
were found available
to be 55power
W. As were found
the figure to be 55 W. As
demonstrates, the the figure
proposed
demonstrates,
MPPT algorithmtheextracts
proposed the MPPT algorithm
real maximum extracts
power fromthe
thereal maximum
module with lowpower from the
oscillation andmodule
small
with low time.
transient oscillation
Figureand15small
depicts transient time. Figure
a case study where 15 depicts
there a case study
is a sudden change where there
in the is radiation
solar a sudden
change
level, in the
where solar
the radiation
proposed MPPT level, where the
technique ableproposed
to track theMPPT technique
maximum ablewith
power to track the maximum
low oscillation and
power with low oscillation and small transient time. A two-step test is performed and the result is
shown in Figure 16, where partial shading conditions are formed by shading a part of the PV module,
then this shading is removed, and as shown in the figure, the proposed fixed-frequency predictive
MPPT effectively tracks the MPP.
The turn-on ZVS is indicated in Figure 17. During turning on the switch, its current should be
radiation level and the maximum monitored available power were found to be 55 W. As the figure
demonstrates, the proposed MPPT algorithm extracts the real maximum power from the module
with low oscillation and small transient time. Figure 15 depicts a case study where there is a sudden
change in the solar radiation level, where the proposed MPPT technique able to track the maximum
power with
Energies 2020,low oscillation and small transient time. A two-step test is performed and the result
13, 1460 is 16
12 of
shown in Figure 16, where partial shading conditions are formed by shading a part of the PV module,
then this shading is removed, and as shown in the figure, the proposed fixed-frequency predictive
smalleffectively
MPPT transient time.
tracksAthe
two-step
MPP. test is performed and the result is shown in Figure 16, where partial
shading conditions are formed
The turn-on ZVS is indicated byin
shading
Figurea17.
part of the turning
During PV module, then
on the this shading
switch, is removed,
its current and
should be
as shown
negative by in the figure,
passing the proposed
through the switchfixed-frequency
body diode. So predictive MPPT
that the switch effectivelyistracks
capacitance the MPP.
discharged and
ZVS is achieved.

Figure 14. Proposed fixed-frequency predictive MPPT operation waveforms under partial shading
Figure 14. Proposed fixed-frequency predictive MPPT operation waveforms under partial shading
condition
Energies 2020, 13,(step
x FORdown
PEERin the irradiation (Ch2 (green) is PV voltage, Ch4 (pink line) is the PV current),12 of 15
REVIEW
condition (step down in the irradiation (Ch2 (green) is PV voltage, Ch4 (pink line) is the PV current),
the PV extracted power (lavender line).
the PV extracted power (lavender line).

Figure
Figure15.15. Proposed fixed-frequencypredictive
Proposed fixed-frequency predictiveMPPT
MPPT operation
operation waveforms
waveforms afterafter removing
removing the
the partial
partial
shadingshading condition
condition (step(step
up inup inirradiation)
the the irradiation)
(Ch2(Ch2 (green)
(green) is PV
is PV voltage,
voltage, Ch4Ch4 (pink
(pink line)
line) is the
is the PV
PV current), the PV extracted power (lavender
current), the PV extracted power (lavender line). line).

Figure
Figure16.16.Proposed
Proposed Fixed-Frequency
Fixed-FrequencyPredictive
PredictiveMPPT
MPPToperation
operationwaveforms
waveformsunder
underpartial
partialshading
shading
condition
condition (two-step step down then step up in the irradiation) (Ch2 (green) is PV voltage,Ch4
(two-step step down then step up in the irradiation) ) (Ch2 (green) is PV voltage, Ch4(pink
(pink
line) is the PV current), the PV extracted power (lavender line
line) is the PV current), the PV extracted power (lavender line).).
Energies 2020, 13, 1460 13 of 16

Figure
The 16. Proposed
turn-on ZVS isFixed-Frequency Predictive
indicated in Figure MPPT turning
17. During operationonwaveforms under
the switch, its partial
current shading
should be
condition
negative (two-step
by passing step down
through then step
the switch up in
body the irradiation)
diode. ) (Ch2
So that the (green)
switch is PV voltage,
capacitance Ch4 (pink and
is discharged
line)
ZVS is is the PV current), the PV extracted power (lavender line ).
achieved.

Figure 17. The turn on ZVS for the bridge.

Figure 17.

6. Conclusions
6. Conclusions
This
This manuscript
manuscripthashaspresented
presenteda fixed frequency
a fixed frequencypredictive-MPPT
predictive-MPPTfor phase-shift modulated
for phase-shift LLC
modulated
resonant converters
LLC resonant for PVfor
converters applications. DesignDesign
PV applications. steps for the control
steps for the algorithm with the with
control algorithm LLC resonant
the LLC
converter have been presented in a comprehensive way. The proposed FFPS predictive
resonant converter have been presented in a comprehensive way. The proposed FFPS predictive MPPT achieved
the
MPPTmaximum
achieved available power available
the maximum from the PV module
power fromwith an efficiency
the PV around
module with 99% and around
an efficiency steady state
99%
oscillation around ±0.5 W @ 115 W. The proposed fixed frequency predictive MPPT
and steady state oscillation around ±0.5 W @ 115 W. The proposed fixed frequency predictivefor LLC resonant
MPPT
converters demonstrated several advantages over other MPPT techniques suitable to be applied to
LLC resonant converters such as: (1) a small steady state error due to the use of a PI controller. (2) Fast
transient performance due to prediction of the reference voltage, and (3) a simple parameter design
due to the usage of fixed frequency modulation. A 300 W simulation model was implemented, then a
250 W hardware step-up was tested. Both simulation and hardware results are consistent and proved
the high performance of the proposed predictive MPPT algorithm for LLC resonant converters.

Author Contributions: O.A.-R., M.A.I., N.A. and M.A., conceived, designed, implemented the proposed control
strategy, and wrote the manuscript. M.O. and R.K. were responsible for guidance and a number of key suggestions.
All authors have read and agreed to the published version of the manuscript.
Funding: This work is supported in part by the Egyptian Science and Technology Development Funds (STDF) and
German Academic Exchange Service (DAAD) under German Egyptian Mobility Program For Scientific Exchange
and Excellence Development (GE-SEED) ID: 30290.
Acknowledgments: This work was supported by the German Research Foundation (DFG) and the Technical
University of Munich (TUM) in the framework of the Open Access Publishing Program. This work is supported
in part by the Egyptian Science and Technology Development Funds (STDF) and German Academic Exchange
Service (DAAD) under German Egyptian Mobility Program For Scientific Exchange and Excellence Development
(GE-SEED) ID: 30290. Any opinions, findings, and conclusions or recommendations expressed in this material are
those of the author(s) and do not necessarily reflect the views of the funding agencies.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Fraunhofer, I.S.E. Photovoltaics Report, Tech. Rep. Nov. 2019. Available online: https://www.ise.fraunhofer.de/
content/dam/ise/de/documents/publications/studies/Photovoltaics-Report.pdf (accessed on 14 November 2019).
2. Jäger-Waldau, A. PV Status Report 2018. Available online: https://publications.jrc.ec.europa.eu/repository/
bitstream/JRC113626/pv_status_report_2018_online.pdf (accessed on 4 November 2018).
3. Global Market Outlook for Solar Power/2018–2022. Available online: https://www.solarpowereurope.org/
wp-content/uploads/2018/09/Global-Market-Outlook-2018-2022.pdf (accessed on 17 November 2017).
Energies 2020, 13, 1460 14 of 16

4. Renewables 2019 Fuel Report—October 2019. Available online: https://www.iea.org/reports/renewables-2019

(accessed on 21 October 2019).
5. Kjaer, S.B.; Pedersen, J.K.; Blaabjerg, F. A review of single-phase grid-connected inverters for photovoltaic
modules. IEEE Trans. Ind. Appl. 2005, 41, 1292–1306. [CrossRef]
6. Abdel-Rahim, O.; Funato, H.; Haruna, J. An efficient and high-gain inverter based on the 3s inverter employs
model predictive control for PV. J. Electr. Eng. Technol. 2017, 12, 1484–1494.
7. Roman, E.; Alonso, R.; Ibanez, P.; Elorduizapatarietxe, S.; Goitia, D. Intelligent PV Module for Grid-Connected
PV Systems. IEEE Trans. Ind. Electron. 2006, 53, 1066–1073. [CrossRef]
8. Parimita, M.; Muneer, T.; KolheSolar, M. Photovoltaic System Applications: A Guidebook for off-Grid Electrification;
Springer: Berlin, Germany, 2015; p. 50.
9. Li, J.; Chen, Y.; Liu, Y. Research on a stand-alone photovoltaic system with a supercapacitor as the energy
storage device. Energy Procedia 2012, 16, 1693–1700. [CrossRef]
10. Merei, G.; Berger, C.; Sauer, D.U. Optimization of an off-grid hybrid PV–Wind–Diesel system with different
battery technologies using genetic algorithm. Sol. Energy 2013, 97, 460–473. [CrossRef]
11. Mohammed, O.; Amirat, Y.; Benbouzid, M.; Elbast, A. Optimal Design of a PV/Fuel Cell Hybrid Power System for
the City of Brest in France; IEEEICGE: Sfax, Tunisia, 2014; pp. 119–123.
12. Kouro, B.W.S.; Abu-Rub, H.; Blaabjerg, F. Power electronics for renewable energy systems, transportation,
and industrial applications. In Photovoltaic Energy Conversion Systems, 1st ed.; Wile: Hoboken, NJ, USA, 2014;
Chart 7.
13. Meinhardt, M.; Cramer, G.; Burger, B.; Zacharias, P. Multi-string-converter with reduced specific costs and
enhanced functionality. Sol. Energy 2001, 69, 217–227. [CrossRef]
14. Yang, Y.; Kim, K.A.; Blaabjerg, F.; Sangwongwanich, A. 1-Introduction, in Advances in Grid-Connected Photovoltaic
Power Conversion Systems; Yang, Y., Kim, K.A., Blaabjerg, F., Sangwongwanich, A., Eds.; Woodhead Publishing:
Cambridge, UK, 2019; pp. 1–13.
15. Ahmed, M.E.; Orabi, M.; AbdelRahim, O.M. Two-stage micro-grid inverter with high-voltage gain for
photovoltaic applications. IET Power Electron. 2013, 6, 1812–1821. [CrossRef]
16. Kouro, S.; Leon, J.I.; Vinnikov, D.; Franquelo, L.G. Grid-Connected Photovoltaic Systems: An Overview of
Recent Research and Emerging PV Converter Technology. IEEE Ind. Electron. Mag. 2015, 9, 47–61. [CrossRef]
17. Chen, L.; Amirahmadi, A.; Zhang, Q.; Kutkut, N.; Batarseh, I. Design and Implementation of Three-Phase
Two-Stage Grid-Connected Module Integrated Converter. IEEE Trans. Power Electron. 2014, 29, 3881–3892.
[CrossRef]
18. Çelik, Ö.; Teke, A.; Tan, A. Overview of micro-inverters as a challenging technology in photovoltaic
applications. Renew. Sustain. Energy Rev. 2018, 82, 3191–3206. [CrossRef]
19. Orabi, M.; Ahmed, M.; Abdel-Rahim, O. A Single-stage High Boosting Ratio Converter for Grid-connected
Photovoltaic Systems. Electr. Power Compon. Syst. 2013, 41, 896–911. [CrossRef]
20. Razi, A.; Hidayat, M.N.; Seroji, M.N. Microinverter Topology based Single-stage Grid-connected Photovoltaic
System: A Review. Indones. J. Electr. Eng. Comput. Sci. 2018, 11, 645–651. [CrossRef]
21. Oriti, G.; Julian, A.L.; Bailey, T.D. PV power conditioning system with LLC resonant converter in DCM. In
Proceedings of the 2014 IEEE Energy Conversion Congress and Exposition (ECCE), Pittsburgh, PA, USA,
14–18 September 2014; pp. 4262–4268.
22. Darwish, A.; Holliday, D.; Ahmed, S.; Massoud, A.M.; Williams, B.W. A Single-Stage Three-Phase Inverter
Based on Cuk Converters for PV Applications. IEEE J. Emerg. Sel. Top. Power Electron. 2014, 2, 797–807.
[CrossRef]
23. Abdel-Rahim, O.; Funato, H.; Haruna, J. A comprehensive study of three high gain dc-dc topologies based
on cockcroft-walton voltage-multiplier for reduced power pv applications IEEJ Transactions on Electrical
and Electronic Engineering(TEEE D). IEEJ Trans. 2018, 13, 642–651.
24. Liang, Z.; Guo, R.; Li, J.; Huang, A.Q. A High-Efficiency PV Module-Integrated DC/DC Converter for PV
Energy Harvest in FREEDM Systems. IEEE Trans. Power Electron. 2011, 26, 897–909. [CrossRef]
25. Jang, S.; Won, C.; Lee, B.; Hur, J. Fuel Cell Generation System With a New Active Clamping Current-Fed
Half-Bridge Converter. IEEE Trans. Energy Convers. 2007, 22, 332–340. [CrossRef]
Energies 2020, 13, 1460 15 of 16

26. Bai, C.; Han, B.; Kim, M. Current-fed dual-half-bridge converter directly connected with half-bridge inverter
for residential photovoltaic system. Sol. Energy 2018, 174, 108–120. [CrossRef]
27. Kasa, N.; Iida, T.; Liang, C. Flyback inverter controlled by sensorless current MPPT for photovoltaic power
system. IEEE Trans. Ind. Electron. 2005, 52, 1145–1152. [CrossRef]
28. Hasan, R.; Mekhilef, S. Highly efficient flyback microinverter for grid-connected rooftop PV system.
Sol. Energy 2017, 146, 511–522. [CrossRef]
29. Geury, T.; Pinto, S.; Gyselinck, J. Direct control method for a PV system integrated in an Indirect Matrix
Converter-based UPQC. In Proceedings of the 2016 IEEE Smart Energy Grid Engineering (SEGE), Oshawa,
ON, Canada, 21–24 August 2016; pp. 117–122.
30. Haque, M.M.; Wolfs, P.; Alahakoon, S. Dual active bridge and matrix converter based three-port converter
topology for grid interactive PV-battery system. In Proceedings of the 2017 Australasian Universities Power
Engineering Conference (AUPEC), Melbourne, VIC, Australia, 19–22 November 2017; pp. 1–6.
31. Yilei, G.; Zhengyu, L.; Zhaoming, Q. Three-level LLC series resonant DC/DC converter. IEEE Trans. Power
Electron. 2004, 20, 781–789.
32. Beiranvand, R.; Rashidian, B.; Zolghadri, M.R.; Alavi, S.M.H. Using LLC Resonant Converter for Designing
Wide-Range Voltage Source. IEEE Trans. Ind. Electron. 2011, 58, 1746–1756. [CrossRef]
33. Harischandrappa, N.; Bhat, A.K.S. A Fixed-Frequency $LCL$-Type Series Resonant Converter With a
Capacitive Output Filter Using a Modified Gating Scheme. IEEE Trans. Ind. Appl. 2014, 50, 4056–4064.
[CrossRef]
34. Cha, W.J.; Kwon, J.M.; Kwon, B.H. Highly efficient step-up DC-DC converter for photovoltaic micro-inverter.
Sol. Energy 2016, 135, 14–21. [CrossRef]
35. Rashid, M.H.; Hui, S.Y.; Chung, H.S.H. 12-Resonant and Soft-Switching Converters. In Power Electronics
Handbook, 4th ed.; Rashid, M.H., Ed.; Butterworth-Heinemann: Oxford, UK, 2018; pp. 339–383.
36. Salem, M.; Jusoh, A.; Idris, N.R.N.; Das, S.H.; Alhamrouni, I. Resonant power converters with respect to
passive storage (LC) elements and control techniques–An overview. Renew. Sustain. Energy Rev. 2018, 91,
504–520. [CrossRef]
37. Abdel-Rahim, O.; Funato, H.; Haruna, J. Novel predictive maximum power point tracking techniques for
photovoltaic applications. J. Power Electron. 2016, 16, 277–286. [CrossRef]
38. Metry, M.; Shadmand, M.B.; Balog, R.S.; Rub, H.A. High-efficiency MPPT by model predictive control
considering load disturbances for photovoltaic applications under dynamic weather condition. In Proceedings
of the IECON 2015-41st Annual Conference of the IEEE Industrial Electronics Society, Yokohama, Japan,
9–12 November 2015; pp. 004092–004095.
39. Metry, M.; Shadmand, M.B.; Balog, R.S.; Rub, H.A. A variable step-size MPPT for sensorless current model
predictive control for photovoltaic systems. In Proceedings of the 2016 IEEE Energy Conversion Congress
and Exposition (ECCE), Milwaukee, WI, USA, 18–22 September 2016; pp. 1–8.
40. Sajadian, S.; Ahmadi, R. Model Predictive-Based Maximum Power Point Tracking for Grid-Tied Photovoltaic
Applications Source Inverter. IEEE Trans. Power Electron. 2016, 31, 7611–7620. [CrossRef]
41. Fang, X. Analysis and Design Optimization of Resonant DC-DC Converters. Doctoral Dissertation, University
of Central Florida, Orlando, FL, USA, 2012.
42. Shahul, A.; Oommen, A.P.; Kuruvilla, J. PMPPT Based Resonant Converter for A Standalone Water Pumping
System. Int. J. Eng. Appl. Sci. Technol. 2017, 2, 148–154.
43. Conesa, A.; Velasco, G.; Martinez, H.; Roman, M. LCLC resonant converter as maximum power point tracker
in PV systems. In Proceedings of the 2009 13th European Conference on Power Electronics and Applications,
Barcelona, Spain, 8–10 September 2009; pp. 1–9.
44. Zhang, Q.; Hu, C.; Chen, L.; Amirahmadi, A.; Kutkut, N.; Shen, Z.J.; Batarseh, I. A Center Point Iteration
MPPT Method With Application on the Frequency-Modulated LLC Microinverter. IEEE Trans. Power Electron.
2014, 29, 1262–1274. [CrossRef]
45. Li, X.; Li, H.; Hu, G. Modeling of a fixed-frequency resonant LLC DC/DC converter with capacitive output
filter. In Proceedings of the 2013 IEEE Energy Conversion Congress and Exposition, Denver, CO, USA,
15–19 September 2013; pp. 5456–5461.
46. Alamir, N.; Ismeil, M.A.; Orabi, M. New MPPT technique using phase-shift modulation for LLC resonant
micro-inverter. In Proceedings of the 2017 Nineteenth International Middle East Power Systems Conference
(MEPCON), Cairo, Egypt, 19–21 December 2017; pp. 1465–1470.
Energies 2020, 13, 1460 16 of 16

47. Abdel-Rahim, O.; Alamir, N.; Orabi, M. Mohamed Ismeil Fixed-frequency phase-shift modulated PV-MPPT
for LLC resonant converters. J. Power Electron. 2019, 20, 279–291.
48. Alamir, N.; Abdel-Rahim, O.; Ismeil, M.; Orabi, M.; Kennel, R. Fixed Frequency Predictive MPPT for
Phase-Shift Modulated LLC Resonant Micro-Inverter. In Proceedings of the 2018 20th European Conference
on Power Electronics and Applications (EPE’18 ECCE Europe), Riga, Latvia, 17–21 September 2018; pp. 1–9.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Copyright of Energies (19961073) is the property of MDPI Publishing and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.
 2016 IEEE PES Asia-Pacific Power and Energy Conference - Xi'an - China
A Novel PV Micro-inverter with Parallel
Decoupling Circuit on DC side
Yong-gao Zhang,Nan Zong ,Junxiang Ge,1ia-bin Kang ,Peng He
East China Jiaotong University, Nanchang 330013, Jiangxi Province, China
Email:294746152@qq.com
Abstract-A novel micro-inverter with power decoupling circuit
on De side for photovoltaic power generation is proposed in this
paper. The micro-inverter contains the basic boost circuit which
is parallel with the current source inverter circuit of
high-frequency, and a decoupling capacitor is added in parallel
with two circuits. Peak current control method is used to
increase the voltage across the capacitor and the voltage ripple,
reduce the capacitance requirements and prolong the Iifetime of
inverter. Five working states of the micro-inverter circuit are
investigated. The corresponding circuit parameters and peak
current control strategy are designed. Simulations of the
proposed circuit are performed with PSIM and the results show
that the proposed topology and control strategy can reduce the
decoupling capacitance requirements.
Keywords-boost; micro-inverter; power decoupling; PSIM
1. INTRODUCTION
Recently, the National Energy Administration of China
issued guidelines of energy development. The guidelines
proposed several policy advice including developing distri-
buted energy, releasing the distributed power construction in
the user-side and encouraging diversified investment. Micro-
inverter plays an indispensable role in distributed generation
system.
PV module output constant power by MPPT (Maximum
Power Point Tracking) control, while the power transmitted
n
to the grid includes double-frequency ripple power -31. So
the energy buffer circuit is needed in the inverter circuit. The
traditional buffer circuit is usually connected a large capacity
electrolytic capacitor in parallel with input terminal of the
inverter. At the same time, in order to ensure the life of the
PV module, the voltage ripple needs to be controlled within
2%. These restrictions that result in the decoupling capacitor
is very large, which can only used electrolytic capacitor as
decoupling capacitor. For a traditional single-phase inverter
topology (Fig.l), with the following ratings: Ppy =100W,
Vin=35V and voltage ripple ~Uc=lV, the minimum value of
the decoupling capacitor is 9.1 mF. The large capacity
electrolytic capacitor is not only large in size but also short in
life, which severely limits the service life of the
micro-inverter. Many scholars have put forward different
solutions to prolong the service life of the micro-inverter
[4-6].
978-1-5090-5417 -6/16/ $ 31.00 ©20 16 IEEE
2399
~ P~
PV Ppv o
OC/AC Grid
Module C dec
Fig.l. Single-phase inverter topology.
A full bridge inverter with power decoupling circUlt IS
proposed in [7], the decoupling circuit is independent of the
topology and operation mode of the inverter. Although the
capacity of decoupling capacitor can be reduced by adding
power decoupling circuit, but the decoupling circuit consists
of two decoupling capacitor in series with a capacitor in
parallel, its control strategy is very complicated. Furthermore,
the circuit requires 6 switches, which will increase the cost of
the circuit.
A three port photovoltaic micro-inverter with apower
decoupling circuit is proposed in [8-9]. One port of the
micro-inverter is connected to the grid, the second port is
used to achieve MPPT control, and the third port is used
to achieve power decoupling. Tt is characterized by the
advantages of fewer components, high efficiency. However
the circuit control algorithm is more complex.
For a micro-inverter of power 100W with 50HzlllOV AC,
the expected capacitance is 46 1lF. The average voltage ofthe
decoupling capacitor is stabilized at 150V, and the capacitor
ripple voltage is 41 V.
A new type ofpower decoupling inverter topology, which
combines the buck-boost circuit with the flyback converter, is
proposed in [10]. This circuit can be regarded as two stages
of power transformation. The first stage is used to meet the
requirement of MPPT and increase DC-link voltage, and the
second stage is used to realize the inverter function. This
circuit has the character of simple circuit and high efficiency.
During each switching cycle process, the inductor is charged
by PV modules. When the charging process is end, the
energy stored in inductor completely transferred into the
transformer's magnetizing inductance and the decoupling
capacitance at the same time, then part of the energy
delivered to the secondary side. For an inverter with 180W,
the expected capacitance is 220IlF. lt is obvious that this
micro-inverter cannot realize to reduce the decoupling
capacitor capacitance and prolong the lifetime of the inverter.
 A flyback inverter topology in series with typical boost
converter was presented in this paper. The decoupling
capacitor is connected in series between flyback inverter and
boost converter, which can be reduced to about 60 /lF by
using peak-current control strategy under the same power
level [11]. The control difference between [10] and this paper
is that all power output of PV module is transferred to the
decoupling capacitor in each switching cycle. Thus, the
voltage across the capacitor is increased, so as to provides
convenience to reduce the value of capacitance. Tt has
advantages of sampie structure, less devices and easier control
and so on. This paper is organized as folIows: the main circuit
topology is introduced in section TI, the detailed operation
principle of the inverter is analyzed in section lll, the
considerations of key design and specific control system are
discussed in section IV, simulation results are given in section
V and section VI is the conclusion ofthis paper.
11. PROPOSED TOPOLOGY
Fig.2 shows configuration of the topology. The circuit
contains two high-frequency switches SI and S2 as weil as
two switches S3 and S4 work at power frequency, three diodes
VD 1,VD 2 and VD 3, flyback trans former T, DC capacitor Cl,
decoupling capacitor C 2 , inductor LI, filter composed of Lf
and Cr- The flyback transformer enables the input and output
to achieve electrical isolation. The circuit can be regarded as
two stage gird-connected inverter, the former is a typical
boost circuit, and the later stage is a current source high
frequency link inverter, which is composed of a flyback
converter. The boost circuit is connected to the PV module
and works in DCM mode. SI is used to realize MPPT and the
work of flyback will not affect the MPPT contro!. The output
power of PV module is converted to 220V150Hz AC by the
flyback inverter, and then transferred to the power grid.
Fig.2 The proposed micro-inverter topology
lll. MODE OF OPERATION
As shown in Fig.2(a), the grid voltage and the output
current are respectively expressed by the following formulas:
(1)
Where Ug is the peak gird voltage, 19 is the peak out
current and 0) is the line angular frequency.
The waveform of output power of PV module Ppv and
inverter output instantaneous power Po are shown in Fig.3 (b).
The relation ofthem could be given as folIows:
2400
UI
P(t)=U
o g
Igsin2(0Jt)=~
2 (2)
As can be seen from (2), there exists a large number of
double-frequency power pulsation in the inverter output
power Po, and we can easily achieve:
(3)
Uc2 is the voltage of decoupling capacitor, as shown in
Fig.3(c). lts maximum value and minimum value are Uc_max
and Uc min respectively. Fig.3(d) shows the transformer
magnetizing inductance current. SI controls the maximum
power tracking of PV modules, and the duty cyc1e of SI is
unchanged at stable time, as shown in Fig.3(e). The duty
ratio of S2 changes along with the output power as shown in
Fig3.(f). The duty ratio of S2 is proportional to the
instantaneous power output of inverter. The duty ratio of S2 is
below its average duty ratio, when the AC side output
instantaneous power is less than the input power of PV
modules. The redundant power from PV modules is stored in
decoupling capacitor C 2 • Decoupling capacitor voltage
increased gradually. The duty ratio of S2 is higher than its
average duty ratio, when the AC side output instantaneous
power is greater than the input power of PV modules.
Decoupling capacitor voltage decreased gradually. The
redundant power stored in decoupling capacitor C2 is
supplement the insufficient power of PV module to grid.
(a) ~=-+-~-.c......::::::!!!o..."o;;;::::::~---=""
(b)
(e)
(d)
nnn nnnn nnnn nnnn ~
O~~~UU~~U-~~Lil~LL~~~~
(e)Sln
OO~~nnnnn~~ ~nnnnn~~ ~
(g)-:::S3'----_ _ _ _ _....I....-_ _ _ _ _ _.~
(h) S'-"4'--_ _ _ _ _--,
~
Fig.3 Key wave1'orms 01' the proposed inverter.
In the working mode of decoupling capacitor storage or
release energy, the inductor current and the driving signal
waveform are shown in Fig.4. Five operation stages are
defined according to the operation state of switches and flow
direction of inductance current during one switching cyc1e.

t to the law of energy conservation:
t (6)
t Combining (3)with (6), the excitation inductance current
Fig.4 Inductor current and drive signals waveform during two switching
cycles
Stage 1 [tlrtl]: SI and S2 are turned on, while S3 and S4
are turned off. As shown in Fig.5 (a), the PV module charges
the inductor. Then the inductor current iLl rises Iinearly from
zero. Meanwhile, the magnetizing inductance is charged by
the decoupling capacitor C2, and its current rises also linearly
from zero. The magnetizing inductance current reaches the
maximum value of ip L2, when S2 is turned off at time t=t1.
Before this time the current of secondary side ir.3 is always
zero. Tgnoring the changes of decoupling capacitor voltage
during a switching cycle, we can obtain,
D
=L
. Uc T
Ip_L2 J1 S
2 (9)
(4)
. ip L2
Ip L3 =---
- n
Where, Uc is decoupling capacitor voltage and n is trans-
former turn ratio.
Stage 2 [t1-t2]: SI and S3 are turned on, while S2 and S4
are turned off, which is shown in Fig.5 (b). The PV module
charges the inductance continuously. The inductance current
reaches the maximum value of ip_Ll, when SI is turned off at
time t=t2. Tt can be expressed as folIows:
(5)
The above shows that the output current of PV module
only relates to the turn-on time of SI. Thus the maximum
power tracking can be achieved by controlling the turn-on
time ofS 1.
During this stage, energy stored in the transformer
primary side will be transferred to the secondary windings,
then the secondary current will be flowed into the grid
through S3 and VD3, and the grid-connected current ig is
2401
negative. The energy stored in the transformer is related to
the excitation inductance L2 and its current iL2. Assuming that
the transformer is ideal, the transformer energy pour into grid
without loss in each cycle, which can be caIculated according
can be caIculated as folIows:
Where, h2Jej is the amplitude of primary reference
current.
The turn on time of S2 meets following formula:
D 2 T S -- L 2 • i{2 (8)
Uc
Stage 3 [tJ-f3]: As shown in Fig.5( c), SI and S4 are turned
on, while S2 and S3 are turned off. This working process is
similar to the state 2. The difference between Stage 3 and
stage 2 is that the grid-connected current i g is positive. The
inductor current iu begin to decline at time t=h. Secondary
inductor current iu reduce to zero at time t=t3. If ignoring
pressure drop in the filter inductor, its fall time can be
expressed as:
Stage 4 [t2-t~]: Only S3 is turned on, as shown in Fig.5 (d).
Energy stored in the inductor LI transferred to C2 by VD I
after SI turned off. The capacitor voltage U c2 increases
gradually until the inductor current i Ll drops to zero. The
output current ix is negative at this stage. When the PV
module and inductor L] which charging to the capacitor C2,
S2 is in off state. Thus the energy out of PV module is
completely transferred to the decoupling capacitor in each
switching cycle. The drop time of in can be caIculated as
follows:
(10)
The energy transfer to L] during one cycle can be
expressed as:
(11)
Stage 5 [t4- T,~]: Four switches are turned off in this stage.
There is no current flowing through S3 or S4, and
iu=iL2=iu=O. Capacitor Cr and inductor Lr keep pouring
energy into the grid, as shown in Fig.5 (e). A switching cycle prior T o/4 cycle to the grid, when the AC side output
end and the tlyback transformer is reset, when Sj turned on instantaneous power is less than the input power of PV
again. modules. Integrating the alternating pulse power, then we can
obtain:
J"'l, VD I S2
U I P
f-8
3Tü
!1E = CP -P)dt=~=~
1
(12)
i

•
Cl ~ pv 0 20] 0]

•
SI 8

_. J ' _ • ..J
Where To is the grid cycle, and w=2nITo.The voltage of
decoupling capacitor decline from maximum value Uc max to
(a) Stage 1
minimum value UC_min. The energy released by decoupling
capacitor can be calculated as folIows:
L
_.J., VD I
1
= -C 2 2
1

CI
SI
M 2 2 CU c - max -U emin)
- (13)

• _ • ..J
Combining (12) with (13), the capacitance of C2 and its
instantaneous voltage value can be calculated.
(b) Stage 2

_.~ L
(14)

Where, L1Uc is the voltage ripple of decoupling capacitor,

(e) Stage 3 Uc_ave is the average value of the decoupling capacitor
voltage.
As shown in Fig.6, the capacitance of decoupling
capacitor is inversely proportional to the average voltage
across it and its voltage ripple at constant input power.
Therefore, increasing the value of U c_ave and LiU c can make
the capacitance decrease. However, with the voltage of Uc_ave
and LiVe increasing, the switch stress will be increasing
(d) Stage 4 accordingly. Tt not only causes more difficult of designing,
but also improves the hardware cost of circuit [8]. A
consideration should be found out to make the balance
between reducing of capacitance and reducing the stress of
switching device .

....1 - !lUc=20V
110-
- !lUc=25V
!lUc =30V
(e) Stage 5 !lUc =40V

Fig.5 Five operating stages during one switehing eyele ofthe proposed
inverter
u 60
IV. DESIGN CONSIDERATION

A. Decoupling capacitance design 40

As can be seen from (2), the output power contains ~--

two parts. There constant average power which expressed as , ,

200 250 300
the first part of (2) and alternating pulse power which shown Uc-ave 0/)
as the second part of (2). C2 transfers the stored energy of Fig.6 Deeoupling eapaeitor (C,) versus the average voltage at different

2402
 ripple level.

B. Design of inductance LI and L 2

Because the inverter operates in DCM, the following
inequality must be satisfied.

CD1 +D2 +D3 +D4)~) <~) (15)

CD1 + D2 + D3 )T). <~) (16)

By substituting (4), (10) and (11) into (15), the range of

inductance LI can be given as follows: Fig.7. System control diagram

T"" V. SIMULATION RESULTS

L < (17)
1 1 2
To verify the theoretical analysis, the micro-inverter with
2Ppv ( - + )
U pv Uc -Upl ' decoupling circuit and its control strategy are simulated by
PSIM software. Table.l lists the key parameters of the
In the same way, substituting (5), (8) and (9) into (16), micro-inverter that are used in the simulation.
then the magnetizing inductance L2 can be calculated as
folIows: TableI. Circuit's Parameters
T~
L 2 < --------:---"------- (18) Input
1_+ ~)2
4P pv ( _ _ Voltage 30/v C2 40/flF
U C _ 111ill U g

C. Contral strategy Grid Voltage 220/v LI 60/flH

Fig.7 shows the control system diagram of the circuit.

The switch control signal of SI is obtained from MPPT Input Power IOO/W L2 200/flH
controller. S2 is controlled by the peak current control
strategy. Primary excitation current is the modulating signal Switching Turns Ration
501kHz
Frequency nl :n2:n3= 1:2:2
ofPWM modulator, the actual sampling current ir.2 is used as
carrier signal. At the beginning of each switching cycle, pulse
Fig.8 shows the key simulation results. The inverter
signal of SI sets RS trigger. Then S2 turns on, the current of generates peak current is about 0.62A, which is very close to
magnetic inductance rises linearly from zero. When the the calculated value. The average voltage across the
excitation current in is greater than its reference current, decoupling capacitor is 200V, and the voltage ripple is
comparator reverses and then re sets the RS trigger, and S2 pulsating at the double-line frequency with 40 V peak to peak.
also turns off. So it can control i L2 track the reference value The envelope of the primary magnetizing inductance current
of half sine wave accurately. Thus this circuit ensures that the is half sine wave, which is converted into sinusoidal envelope
output current is a standard sine wave. The total harmonic by S3 and S4, then converted into a standard sine wave pump
distortion (THD) can be reduced by PLL(phase-locked loop). into grid. As shown in Fig.9, iT.3 is the current flowing
PLL circuit generates a sinusoidal reference signal which is through the secondary center-tapped.
consistent with the grid voltage to ensure the grid-connected
current and the grid voltage have the same frequency and
phase. The driving signals of S4 and S3 are obtained by the
.:
comparison of sine reference signal with zero.
In order to control the decoupling capacitor voltage
within the range of design, the voltage control outer loop is
added. Comparing the detected voltage signal Uc with a
given reference value Uc ref, it puts the obtained error signals
into PI controller. Then PI controller outputs an auxiliary
regulator current i*, which will be added to the amplitude
value ofthe reference current in primary inductance to adjust
the instantaneous reference current ir.2Jej;
n I' )
The input current reference lL2Jej can be obtained by Fig.8. Key waveforms in simulation
MPPT while the primary current reference ir.2Jej IS a
time-dependent variable as expressed in (7).

2403

Fig.9. Current through primary and secondary center-tapped
;"~ 0'~ ß ß ß ß
..... .
0; J
' HJ
....... ...... ....... TIme (s)
....,
. [5] Z. Zhang, M. Chen, W. Chen, C. Jiang, and Z. Qian, "Analysis
F ig.lO. Expanded view tor current wavetonns
Fig.10 shows the current flow through LI, primary and
secondary winding. If the ambient temperature and light
conditions are kept constant, the duty cycle of SI will remain
constant, and then iLl is constant. The duty cycle of S2 is
changing as sine wave according to the grid voltage phase,
and the excitation inductor peak current ip L2 is changing with
the duty cycle of S2.
VI. CONCLUSION
A novel micro-inverter with parallel decoupling circuit on
DC side is proposed. During a switching cycle, all power
output of PV module is transferred to the decoupling
capacitor by optimizing the inverter switching logic.
Therefore, it is easy to control the voltage across capacitor
increased, the decoupling capacitance requirements can be
significantly reduced, and a small film capacitor can be
adopted to replace the electrolytic capacitor. Hence, the
lifetime of mic-inverter would be increased. Flyback
trans former is operating in DCM mode, which has
characteristics of small size and high efficiency energy
conversion. The inverter has the advantages of simple
structure, fewer device, easy control and high conversion
efficiency. Simulation results verify the theoretical analysis is
correct.
ACKNOWLEDGMENT
This work is supported by National Natural Science
Foundation of China (51467006); and by Jiangxi provincial
Department of Science and Technology (20151BBE50118,
20161 BBH80032).
References
[1] H. Hu, S. Harb, J. Shen, and I. Batarseh, "A review of power
decoupling techniques for microinverters with three different
decoupling capacitor locations in PV systems," IEEE Trans.
Power Electron., vol. 28, no. 6, pp. 2711 - 2726, Jun. 2013.
[2] GAO Wen-xiang, WANG Ming-yu, WANG Li-jian, LIU
Yang."Review of research on photovoltaic micro-inverter,"
Power System Protection and Control,2012(21):147-155
[3] Li Q, Wolfs P." A Review of the Single Phase Photovoltaic
Module Integrated Converter Topologies With Three Different
DC Link Contigurations," IEEE Transactions on Power
Electronics, 2008, 23(3): 1320-1333.
[4] Chen,Z.,Li, M.,Wu, Q, & Xu, y'''A single-stage grid-connected
flyback inverter with power decoupling function." Applied
':J
Power Electronics Conference and Exposition IEEE, 2015.
and implementation of phase synchronization control strategies
for BCM interleaved flyback micro-inverters," IEEE Trans.
Power Electron., vol. 29, no. 11, pp. 5921-5932, Nov. 2014
[6] G. H. Tan, J. Z. Wang, and Y. C. Ji, "Soft-switching flyback
inverter with enhanced power decoupling for photovoltaic
applications," lET Trans.Elect. Power Appl., vol. I, no. 2, pp.
264-274. Mar. 2007.
[7] Yang, X., Jiang, L., Feng, .I., Sun, c., & Liang, H. "A New
Power Decoupling Circuit for Photovoltaic Inverter." Transac-
tions of China Electrotechnical Society, 30( 16):42 -48.(2015).
[8] H. Hu, S. Harb, N. H. Kutkut, Z. J. Shen, and I. Batarseh, "A
single-stage micro-inverter without using electrolytic
capacitors," IEEE Trans. Power Electron., vol. 28, no. 6, pp.
2677-2687, Jun. 2013.
[9] RU Haibing, HUANG Xiaobo, WANG Wanbao, WU Hongfei,
XING Yan."A PV Micro-inverter Based on Three-port Flyback
With Power Decoupling Function," Proceedings of the CSEE,
33(12):47-54,Apr.20 13.
[10] Z. Chao, H. Xiangning, and Z. Dean,"Design and control of a
novel module integrated converter with power pulsation
decoupling for photovoltaic system,"in Proc. Int. Conf. Electr.
Mach. Syst., 2008, pp. 2637-2639.
[lI] Gonz61ez, L. G., Vanegas, P., Sempertegui, R., & Carranza,
O."Peak control current in boundary conduction mode and
discontinuous conduction mode for inverter with flyback
topology." Power Electronics and Power Quality
Applications IEEE, 2015.
2404
 Three-Port Flyback-Type Single-Phase Micro-
Inverter With Active Power Decoupling Circuit
Yaow-Ming Chen and Chein-Yao Liao
Electric Energy Processing Research Lab. (EEPro)
Department of Electrical Engineering
National Taiwan University
No. 1, Sec. 4, Roosevelt Rd., Taipei, Taiwan
f98921088@ntu.edu.tw
Abstract—The objective of this paper is to propose a novel
three-port flyback-type single-phase micro-inverter with active
power decoupling circuit (APDC). The proposed micro-
inverter can improve the performance of the maximum power
point tracking (MPPT) while reducing the size of the input
capacitor. The electrolytic capacitor would be replaced by the
film ones and the reliability of the micro-inverter can be
improved significantly. The operation principle of the
proposed micro-inverter will be explained. The computer
simulation results of a 200W micro-inverter would be
presented to verify the performance of the proposed micro-
inverter.
Index Terms—Micro-inverter, power decoupling.
I. INTRODUCTION
For the grid-connected PV power system, the dc-ac
inverter can be categorized as: centralized inverter, sting
inverter, multi-string inverter, and micro-inverter. Among
them, the micro-inverter has the advantage of high power
expandability, less shading effect, and easy monitoring. It
had attracted many attentions for PV system development
[1]-[7].
For a single-phase grid-connected micro-inverter PV
system, the output power is composed of a dc component,
PDC, and an ac one, PAC. In general, the PDC is the desired dc
input power from the PV panel, and the PAC must be
decoupled by energy storage components, such as capacitors
or inductors. Otherwise, the efficiency of the maximum
power point tracking (MPPT) will be greatly reduced.
Usually, the way to decouple the PAC is to put a large
electrolytic capacitor in parallel with the PV panel. However,
the large electrolytic capacitor has disadvantages of large
volume, short lifetime and low reliability. Therefore,
different active power decoupling circuits (ADPCs) have
been developed in recent years in order to replace the large
electrolytic capacitor with the small film capacitor.
Many APDCs have been proposed recently. A flyback-
type inverter with APDC is proposed in [2], but the
switching loss and the leakage loss of the inverter are large
978-1-4577-0541-0/11/$26.00 ©2011 IEEE 501
due to the discontinuous conduction mode (DCM) operation
and large peak current. Although the leakage loss is recycled
into the power decoupling capacitor as presented in [3], the
efficiency of the micro-inverter is only increased a little due
to energy double transformation between energy storage
components and large current stress on switches.
The objective of this paper is to propose a novel three-
port flyback-type single phase micro-inverter with APDC to
improve the MPPT performance without using electrolytic
capacitors.
II. CIRCUIT CONFIGURATION OF THE INVERTER
A. Circuit describtion
The proposed micro-inverter with the ADPC is shown in
Fig.1. The power of the PV panel would be delivered by a
flyback converter and an unfolder to grid, and the micro-
inverter would be operated in DCM with a constant
switching frequency. In this system, the on-time of the SM1 is
determined by MPPT control, and the average current at the
primary-side of the transformer (T1) is constant if the sun
illumination is fixed and current flowing out the PV panel
(IPV) is almost constant. Therefore, the electrolytic capacitor
could be replaced by a small film capacitor to filter out the
constant high-frequency current (IMP) at the primary-side of
T1. At the secondary-side of T1, the switching-frequency
component in current (IM2) would be filtered by the LC filter
(Lo andC2), and the sinusoidal current (IAC), which will be in
phase with voltage of the grid (VAC), would be obtained by
the unfolder. Therefore, a sinusoidal real power is produced
and injected to the grid. In order to balance the difference
between the instantaneous output power and the input power,
the ADPC in the power-decoupling-side of the transformer is
necessary.
B. Circuit operation with APDC
The conceptual operation modes of the proposed micro-
inverter are shown in Fig. 2 (a) and (b). The power flow of
the inverter is corresponded to the instantaneous power
waveforms shown in Fig .3. In mode I, when the
instantaneous output power Pout(t) is less than the average
output power (PAC,av) which is equal to average input power
(Pin), the power below the instantaneous output power (blue
area in Fig. 2) would be transferred to grid from Lm, and the
difference between Pout(t) and Pin (green area in Fig. 2) is
stored in Cx. In mode II, as the Pout(t) is larger than Pin, the
power in Cx would be transferred to Lm first, and then the
total power in Lm including input power and the power in Cx
is delivered to grid (purple area and pink area in Fig. 2). The
operational circuit diagrams of different switching stage in
mode I and II are shown in Fig. 4 (a)~(d), and the operation
modes would be explained as follows.
Figure 1. The topology of the proposed micro-inverter.
(a) Mode I (b) Mode II
Figure 2. The operation modes of the micro-inverter.
(b)
(a)
(d)
Figure 4. The switching stages of the Mode I and Mode II.
502
Figure 3. The instantaneous output power.
C. Operation modes
Mode I :
The power is the green and blue areas shown in Fig. 3
while the switching stages are (a), (b), and (c) shown in Fig.
4. In Fig. 4 (a), SM1 is on and the maximum power of the PV
panel is conveyed to Lm. Then, SM1 is off while SM3 and SM5
are on as shown in Fig. 4(b). The stored energy in Lm is
delivered to capacitor Cx which is indicated in the green area
in Fig. 3. Finally, in Fig. 4 (c), SM3 and SM5 are off and SM2 is
on. The rest of power in Lm, the blue area in Fig. 3, is
transferred to grid .
Mode II :
The power is in the purple and pink areas shown in Fig. 3,
while the switching stage are (a), (d), and (c) shown in Fig. 4.
In Fig. 4 (a), SM1 is on and the maximum power of the PV
panel is transferred to Lm. Next, SM1 is off while SM4 and SM6
are on, as shown in Fig.4 (d). The stored energy in Cx, as
shown in the purple area in Fig. 3, is delivered to Lm. Finally,
in Fig. 4 (c), SM4 and SM6 are off and SM2 is on. The total
stored energy in Lm including the energy from the PV panel,
pink area in Fig. 3, and the energy stored in Cx, purple area
in Fig.3, would be transferred to the grid.
(c)
 III. OPERATION PRINCIPLE constant, too. Eventually, the average current IDC flowing
out the PV panel becomes constant and can be expressed as:
The ADPC is introduced in the power-decoupling-side of
the transformer for the proposed three-port micro-inverter. In D1Tsi M1p L mi 2M1p
addition to the VCx, the slope of the iLm would be related with I DC = I M1 = = (3)
the turn ratio of the primary winding and power-decoupling 2Ts 2VDC Ts
winding (Nx/Np). 2) Interval
Fig. 5 shows the operation waveform of the proposed During this time interval, the PDC is larger than PAC. As
micro-inverter. The constant switching period, Ts, includes shown in Fig. 4(b), both switches SM1 and SM2 are off.
the switching stages either in Mode I and II. The operation Switches SM4 and SM6 are turned on to extract energy stored
principle during each interval would be explained as follows. in Lm into the decoupling capacitor, Cx, and the iLm is
discharged by the voltage VCx. Therefore, if the VCx is
1) Interval
constant in one switching period, the iMX could be expressed
As shown in Fig. 4(a), the main switch SM1 is turned on,
as:
and other switches in proposed circuit are turned off. The
main switch current iM1 is equal to the magnetizing inductor
Np Np VCx
current, which can be derived as: i MX ( t ) = i M1p − ( )2 ( t − T1 ) (4)
Nx Nx Lm
VDC
i M1 ( t ) = ( t − T0 ) (1) where Np/Nx are the turns ratio of the primary side and the
Lm
decoupling side of the transformer.
The on-time of the SM1 (D T T T ) is determined When the iMx(t) is decreased to the requested value, the
by the MPPT control, and the peak current flowing through dash line iMXp in Fig.5, switches SM4 and SM6 should be
the SM1 is obtained. turned off. The total time of this interval can be obtained by
the duty ratio D2 of SM4 and SM6.The duration of T2-T1 can
VDC V be determined.
i M1p = i M1 (T1 ) = (T1 − T0 ) = DC D1Ts (2)
Lm Lm
N x 2 Lm Np
D 2 Ts = ( ) ( i M1p − i MXp ) (5)
N p VCx N x

′
3) Interval
During this interval, the switching stages is shown in
Fig. 4(d) where the PDC is smaller than PAC. Switches SM1
and SM2 are turned off but switches SM3 and SM5 are
turned on to transfer part of energy in Cx to Lm, and the iLm
is charged by VCx. If the VCx is constant in one switching
period, the iMX could be expressed as follows.

Np VCx Np
i MX ( t ) = ( )2 ( t − T1 ) + i M1p (6)
Nx Lm Nx

When the iMX reaches the required value, iMXp, switches

SM3 and SM5 should be turned off. Similarly, the duration of
T2-T1 can be determined as following:

N x 2 Lm Np
D 2 Ts = ( ) (i MXp − i M1p ) (7)
N p VCx Nx

Figure 5. Operation waveform of the micro-inverter.
Due to the constant on-time (D1Ts) on SM1 under DCM
operation and the constant input voltage VDC, the iM1p is
503
4) Interval
After switches SM3 ~ SM6 are turned off, the SM2 is
turned on in this interval. The switching stage is shown in
Fig. 4(c). The energy inside the Lm will be transferred to
grid, and the iLm is discharged by the grid voltage, | | .
Since the ac mains voltage is almost constant during the
switching period, the current of SM2, iM2, could be expressed power decoupling function. To derive the capacitance of Cx,
as follows. the input power PDC, the instantaneous output power PAC and
Np V the energy in Cx is required.
N
i M 2 ( t ) = x i Mxp − ( ) 2 AC ( t − T2 ) (8) The dc input power PDC and ac fluctuated output power
Ns Ns Lm
with twice grid frequency could be expressed as follows.
where VAC VAC, |sinωt| .The currents iLm and iM2 are
PDC = VDC I DC (12)
decreased to zero at T3. The duration of this interval can be
expressed as : 1
PAC = VACI AC = VAC, pk I AC, pk (1 − cos 2ωt ) (13)
2
N x Ns 2 Lm
D 3Ts = T3 − T2 = ( ) i MXp (9) where VAC VAC, |sinωt| and IAC IAC, |sinωt|.
N s N p VAC
The capacitor voltage VCx, the average current of iMX,
After the LC filter, the switching-frequency component and the instantaneous power of Cx could be expressed as:
of the iM2 will be filtered out and the average current, which
is equal to the injected ac current, can be obtained: VCx = VCx ( dc ) + VCx , pk sin ωt (14)

2
1 T3 N x 2 LiMXp dVCx ( dc )
I AC = I M 2 =
Ts ∫
T2
iM 2 (t ) dt =( )
N p 2TsV AC
I MX = C x
dt
= 2ωC x VCx cos 2ωt (15)

2
N x 2 LI MXp PCx = VCx I MX
=( ) sin ωt (10)
N p 2Ts VAC, pk
= 2ωC x VCx , pk (VCx ( dc) + VCx , pk sin ωt ) cos 2ωt (16)
Where iMX IMX |sinωt|.
When the PAC is smaller than the PDC as the green area
The above derived current equation is a rectified indicated Fig.3, the rest of power would be transferred into
sinusoidal function. Therefore, the unfolder which consists Cx. Hence, the total energy difference between PDC and PAC
of SM7~SM10 are used to change the direction of the injected would be delivered into Cx, and the magnitude of the VCx,pk
current into the ac mains. Finally, the injected ac output can be determined as:
current of the proposed micro-inverter can be obtained:
TAC TAC

I AC = I AC, pk sin ωt = (
Nx 2
)
LI 2MXp
sin ωt (11) ∫0
8 PCx dt = ∫0
8 Pdc − PAC dt (17)
N p 2Ts VAC, pk
2 VAC I AC TAC
5) Interval VCx , pk = −VCx ( dc ) + VCx ( dc ) + (18)
4πC x
All the switches are off because of the DCM operation
of the proposed micro-inverter.
where TAC is the grid frequency.
IV. THE DECOUPLING CAPACITOR SELECTION From (18), the ripple voltage on Cx can be reduced as the
In section II, the mode operation of the proposed micro- dc component of the VCx and the Cx are larger. In order to
inverter is described, and the constant input voltage VDC and discharge and charge the iLm, the VCx,pk should be smaller
current IDC due to the fixed input power PDC is obtained with than VCx(dc), and the minimum value of VCx to make the
the APDC and its control strategy. Then, the difference micro-inverter operated in the DCM can be expressed as:
between Pdc and Pac would appear on the decoupling
capacitor (Cx). However, the large ripple voltage on the Cx
would not affect the MPPT control and the THD of the ac 2 VAC I ACTAC
VCx , min = 2VCx ( dc) − VCx ( dc ) + (19)
output current if the small Cx is used. Therefore, in addition 4πC x
to the input capacitor (CDC), the Cx can used film capacitor to
replace the large electrolytic capacitor, and the lifetime of the V. THE MAGNETIZING INDUCTOR SELECTION AND
micro-inverter can be increased. CURRENT CONTROL DESIGN
Although the large voltage variation on Cx is permitted, Because the inverter must be operated in DCM, the sum
the magnetizing current is needed to be charge and discharge of the time intervals of 1), 2), and 4) or 1),3), and 4) should
by high enough voltage across Cx (VCx) in DCM operation. be shorter than one switching period, Ts. Hence, the
Then, the Cx needs to be carefully chosen to achieve the following inequality must be satisfied.

504
 D1Ts + D 2 Ts + D 3Ts < Ts
or D1Ts + D '2 Ts + D 3Ts < Ts
Before the inequality of (20) is solved, the iMXp should be
determined first. By arranging the equality of the PDC and the
average value of PAC, the iM1p can be obtained. Then, by
plugging iM1p into (10), the iMXp could be obtained:
VAC, pk I AC, pk Ts
i M1p =
Lm
Np
i MXp = 2 i M1p sin ωt
Nx
By substituting (2),(5) or (7), (9), and (22) into (20), the
maximum magnetizing (Lm) can be calculated.
Ts VDC VAC,pk VCx
Lm ≤ ( )2
VAC I AC Nx
VAC,pk VCx + VDC VAC,pk 1 − 2 sin ωt + 2 s VDC VCx
Np
The minimum value of the right side of (23) would
happen as the minimum capacitor voltage VCx,min is achieved
at ωt 7π/4. Therefore, (23) could be rewritten as (24).
Ts VDC VAC,pk VCx ,min
Lm ≤ ( )2
VAC I AC N N
VAC,pk VCx ,min + 2 x VDC VAC,pk + 2 s VDC VCx ,min
Np Np
From (24), the VCx,min is related with the turns ratio,
Nx/Np, under fixed turns ratio, Ns/Np, and maximum Lm. As
the Nx/Np becomes smaller, the VCx can be smaller, too.
Therefore, the voltage stress of switches in APDC and the
size of Cx can be easily designed under smaller turn ratio
Nx/Np.
After the circuit parameters are determined, the switches
control design will be implemented. The control block
diagram of the proposed micro-inverter is shown in Fig. 6.
At the upper left part of Fig.6, the VDC and IDC are sensed by
MPPT control, then the current commend IDC* can be
generated. If the maximum power of the PV panel varies, the
IDC* will be changed and the actual current, IDC, would track
the maximum power of the PV panel. The on-time of the SM1
would almost be constant under steady state. After SM1 is
turned off, the NLT (Negative Ledge Trigger) would be
conveyed a pulse to S-R latch to turn on the SM4 and SM6 or
SM3 and SM5.
By comparing the iM1p* and iMXp*, one of the diagonal
pair switches in APDC are turned on, and the current, iMXp,
would be charged or discharged by Cx. When the current,
iMXp, reaches the current commend, iMXp*, the SM4 and SM6 or
SM3 and SM5 should be turned off by a pulse triggering the
reset function of the S-R latch. Finally, the SM2 would be
turned on until the next switching period begins. The
comparator in the bottom of the Fig. 6 is to get the polarity of
the VAC, and the SM8 and SM10 or SM7 and SM9 are triggered
(20) by the comparator. Therefore, the output current would be in
phase with the grid voltage and the pure real power is
obtained.
(21)
(22)
Figure 6. The control block diagram of the presented inverter.
N
VI. COMPUTER SIMULATION RESULTS
Np
A prototype circuit has been designed to verify the
(23)
performance of the three-port flyback-type micro-inverter.
The circuit parameters of the prototype circuit are tabulated
in Table I.
Fig. 7 (a) and (b) are the computer simulation results of
the proposed micro-inverter without and with the APDC,
respectively. If the APDC and large input capacitor are not
included in the proposed micro-inverter, the sinusoidal ripple
would appear on the terminal voltage VDC and current IDC of
(24) the PV panel due to the sinusoidal ac output power.
Therefore, the output power of the PV panel will be reduced
because of the large variation on the PV terminal voltage.
Also, the total harmonic distortion (THD) of the output
current will be increased.
On the other hand, the PV terminal voltage will be almost
constant because of the steady current controlled by the SM1
in the primary-side of the transformer. Therefore, the
performance of the MPPT function and the output current
THD will be improved. Eventually, an input electrolytic
capacitor can be replaced by a film capacitor and the
reliability of the micro-inverter might be increased greatly.
VII. CONCLUSIONS
The objective of this paper is to propose a novel three-
port flyback-type micro-inverter with an APDC. The
proposed micro-inverter can improve the performance of the
MPPT while reducing the size of the input capacitor. The
electrolytic capacitor can be replaced by the film one and the
reliability of the micro-inverter can be improved greatly. The
operation principle and circuit design of the proposed micro-
inverter is introduced in this paper. The computer simulation
results show that the input power of the proposed micro-
inverter with the APDC and film capacitors is almost
constant with very good MPPT performance.
505
 References
[1] Haibing Hu, S. Harb, N. Kutkut, I. Batarseh, and Z.J. Shen , "Power
decoupling techniques for micro-inverters in PV systems-a review," VDC
IEEE Energy Conversion Congress and Exposition, pp.3235-3240,
2010.
[2] 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 Transactions on IDC
Power Electronics, vol.21, no.5, pp.1264-1272, Sept. 2006.
[3] S.B. Kjaer, and F. Blaabjerg, "Design optimization of a single phase
inverter for photovoltaic applications," IEEE Power Electronics
Specialist Conference, pp. 1183- 1190, 2003.
[4] T. Shimizu and S. Suzuki, "A single-phase grid-connected inverter VAC
with power decoupling function," International Power Electronics
Conference, pp.2918-2923, 2010.
[5] G. H. Tan, J. Z. Wang, and Y. C. Ji, "Soft-switching flyback inverter
with enhanced power decoupling for photovoltaic applications," IET
Transactions on Electric Power Applications, vol. 1, no. 2, pp.264- IAC
274, March 2007.
[6] Zhang Chao, He Xiangning, and Zhao Dean, "Design and control of a
novel module integrated converter with power pulsation decoupling
for photovoltaic system," International Conference on Electrical
Machines and Systems, pp.2637-2639, 2008. (a) Without the APDC
[7] A.C. Kyritsis, N.P. Papanikolaou, and E.C. Tatakis, "Enhanced current
pulsation smoothing parallel active filter for single stage grid-
connected AC-PV modules," European Power Electronics and Motion
Control Conference, pp.1287-1292, 2008.

VDC
TABLE I. CIRCUIT PARAMETERS OF THE PROPOSED MICRO-INVERTER

IDC

VAC

IAC

VCx

(b) With the APDC

Figure 7. Computer simulations of the micro-inverter with and without
the APDC.

506