DC–DC converter without electrolytic capacitors for renewable

advertisement
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
High Step-up Converter without Electrolytic
Capacitors for Renewable Energy Applications
Yihua Hu, Member, IEEE, Wenping Cao, Senior Member, IEEE, Bing Ji, Member, IEEE, Xiangning
He, Fellow, IEEE, James Kirtley, Fellow, IEEE

Abstract—Traditional electrolytic capacitors used in power
converters are bulky, unreliable and short in their service life.
This paper proposes the use of film capacitors in a high step-up
DC-DC converter for renewable energy applications including
photovoltaics and fuel cells. The converter is of a three-phase
modular structure and six pulse rectifiers are employed to reduce
the output voltage ripple. The primary side of the converter
includes three interleaved inductors, three main switches and an
active clamp circuit. Therefore, the input capacitor can be
eliminated and the input current is greatly reduced. The zero
voltage switching is achieved during the switching transition for
all the active switches so that the switching losses can be reduced
greatly. Magnetic energy stored in the leakage inductance is
recovered and the reverse-recovery issue associated with diodes is
effectively addressed by the leakage inductance of a built-in
transformer. The proposed converter is justified by simulation and
experimental tests on a 1000W prototype.
Index Terms—DC-DC converter, electrolytic capacitor, fuel
cells, modularization, photovoltaics, power loss.
I. INTRODUCTION
R
ENEWABLE energy is high on the international and
national agendas. Photovoltaics (PVs) and fuel cells are
two prominent examples and their output is typically low
voltage direct current (DC). In these applications, high step-up
DC-DC converters with low cost and high reliability are highly
desired [1]-[5]. However, large electrolytic capacitors are
generally used at the front-end and the output terminal of the
converter [2], in order to minimize the current and voltage
ripples. These capacitors require galvanic isolation [7] and
provide only a limited service life, especially in the PV microinverter at outdoor conditions [6].
Voltage gain is also a key performance indicator for DCDC converters. In the literature, a variety of isolated high stepup topologies to increase the voltage gain are reported [8]-[14].
Among these is an interesting concept of isolating inter-cell
transformer and interleaved technology [14]. Nonetheless, due
to its multi-cell structure, the bridge circuit suffers from a
breakthrough risk and the voltage stress of switching devices is
high on the high voltage side. In [15], a group of interleaved
Flyback converters with three winding coupled inductors are
developed but their current ripples are also high. Thereby new
topologies are proposed in [16]-[17] to lower input current
ripples while still achieving a high voltage gain. Their
***
drawback lies in a need for an input capacitor (small) and an
output capacitor (large). On the contrary, a symmetrical
rectifier configuration is used [18] to decrease the output
current ripple but the current ripple on the primary side
structure is high. This work is set out to achieve the high
voltage gain and low output ripples without the use of
electrolytic capacitors, by proposing a new topology which
combines the features of electrical isolation and modular
structure so as to improve the reliability and cost-effectiveness.
II. PRINCIPLE OF OPERATION
The proposed three-phase interleaved converter without
electrolytic capacitors is presented in Fig. 1, where S1, S2, S3
are the main switch devices, Sc1, Sc2 and Sc3 are clamp switches.
The switch devices also form a three-phase bridge.
The capacitor Cc and the switches Sxxx are employed to
clamp the switch devices and to realize soft switching. For the
three coupled inductors L1, L2 and L3, their primary windings
are interleaved connected with each other and their secondary
windings are star connected for a high output voltage gain. The
winding turns of the coupled inductors are represented by n1
(primary) and n2 (secondary), respectively, and the turns ratio N
is derived from n2/n1. The coupling reference of L1, L2 and L3
are marked with “*”, “○” and “□”, respectively. LLK1, LLK2 and
LLK3 denote the leakage inductances of the coupled inductors;
Do1-Do6 are the rectifier diodes; Co is the output filter capacitor.
There are six operating stages during one switching period.
The equivalent circuits for each stage are presented in Fig. 3
and explained in detail as follows.
L3
*
Vin
Sc3
L3a
n1 L2
L2a
n1 L1
Sc2
Sc1
S3
S2
S1
L1b
n2
n2
L3b
Cc
Do1
LLk1
L1a
n1
Do3 Do5
L2b n2
Co
LLk2
R V
out
*
LLk3
Do2
Do4 Do6
Fig. 1 The proposed DC-DC converter without electrolytic capacitors.
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
L3a
n1
L3
*
Vin
Sc3
L2
L2a
n1
L1
L1b
n2
n2
L3b
Cc
S3
LLk1
L1a
n1
Sc1
Sc2
S1
S2
Do1
Do3 Do5
L2b n2
Co
R
Vout
Co
R
Vout
LLk2
*
LLk3
Do2
Do4 Do6
Do1
Do3 Do5
(a) Stage 1 [t0~t1]
L3
*
Vin
L3a
n1
Sc3
L2
L2a
n1
L1
Sc2
Sc1
S2
S1
L1b
n2
n2
L3b
Cc
S3
LLk1
L1a
n1
L2b n2
LLk2
*
LLk3
Do2
Do4 Do6
Do1
Do3 Do5
(b) Stage 2 [t1~t2]
L3
*
Vin
L3a
n1
L2
L2a
n1
L1
Sc3
Sc2
Sc1
S3
S2
S1
LLk1
L1a
n1
L1b
n2
n2
L3b
Cc
L2b n2
Co
LLk2
R
Vout
*
LLk3
Do2
Do4 Do6
(c) Stage 3 [t2~t3]
L3
*
Vin
L3a
n1
L2
L2a
n1
L1
Sc3
Sc2
Sc1
S3
S2
S1
L1b
n2
n2
L3b
Cc
Do1
LLk1
L1a
n1
Do3 Do5
*
Vin
L3a
n1
Co
LLk2
R
Vout
*
LLk3
L2
L2a
n1
L1
Sc3
Sc2
Sc1
Do2
S3
S2
S1
Do1
LLk1
L1a
n1
L1b
n2
Do3 Do5
Co
R
Vout
Co
R
Vout
*
Do2
Do4 Do6
Do1
Do3 Do5
(e) Stage 5 [t4~t5]
L3
*
Vin
Sc3
L3a
n1
Sc2
L2
L2a
n1
Sc1
Cc
S3
S2
L1
S1
LLk1
L1a
n1
L1b
n2
n2
L3b
L2b n2
LLk2
*
LLk3
Do2
Do4 Do6
(f) Stage 6 [t5~t6]
Fig. 3 Six operating stages of the proposed converter.
Vin
 Vout
1 D
(t  t1 )
LLK 1
(1)
Vin
 Vout
(2)
1
D
iLK 2  iLK 2(t2 ) 
(t  t1 )
LLK 2
By Kirchhoff’s current law (KCL), the current on the
secondary side of L3 is given by (rewrite i3=)
(3)
iLK 3  iLK 1  iLK 2  0
LLk2
LLk3
iLK 1  
N
N
Do4 Do6
L2b n2
n2
L3b
Cc
B. Stage 2
In this stage, the main switches S1 and S3 turn on and the
active clamp switch Sc2 is in on state. At t1, the current in the
primary of L1 and L3 increases whilst that of L2 falls below zero.
During this stage, L1 and L2 work in forward mode, L3 works in
flyback mode, and the rectifier diodes D02, D04 and D05 conduct.
The secondary windings of L1 and L3 are series connected to
charge the output for high voltage gain; these of L2 and L3 are
also series connected to charge the output for lowering the
output voltage ripple.
The current in the secondary windings of L1 and L2 can be
expressed by Eqs. 1 and 2.
L2b n2
(d) Stage 4 [t3~t4]
L3
A. Stage 1
Before t0, the main switch S1 and the clamp Switch Sc1 are
in off state and the common clamp capacitor Cc is resonant with
the leakage inductance of L1. From t0 onwards, the voltage of
Sc1 increases from zero and zero voltage switching (ZVS) is
realized on the clamp switch Sc1. The turn-on signal of the main
switch S1 is given in this period when its parasitic diode is in on
state, and S1 turns on with ZVS. Meanwhile, the main switch S2
turns off and its parasitic capacitor Cs2 is charged by the
magnetizing current of L2. Due to its parasitic capacitance, the
main switch S2 turns off with ZVS. In this period, the switching
voltage of the main switch S2 reaches the level of Cc and the
parasitic diode of Sc2 conducts. The voltage of the main switch
S2 is clamped to the voltage of Cc. The primary-side current of
L2 begins to decrease and that of L1 begins to increase. L2 and
L3 work in forward condition and L1 in flyback condition.
Corresponding rectifier diodes D01, D04 and D06 conduct.
C. Stage 3
At t2, Sc2 turns off and the clamp capacitor Cs2 is resonant
with the leakage inductance of L2. While the voltage across the
parasitic capacitance of the main switch decreases, the voltage
of Sc2 increases from zero, and ZVS is realized on Sc2. The
turn-on signal of S2 is given when its parasitic diode is in on
state and S2 turns on with ZVS. The main switch S3 turns off,
and then its parasitic capacitor Cs3 is charged by the
magnetizing current of L3. In this period, the voltage of the
main switch S3 reaches the voltage of Cc and that of the main
switch S3 is clamped to the voltage of Cc3. L1 and L2 work in
forward mode and L3 works in flyback mode. The antiparallel
diode of Sc3 and D02, D04 and D05 all conduct. The secondary
windings of L1 and L3 are series connected to charge the output
for a high voltage gain and these of L2 and L3 also series
connected to charge the output and to reduce the output voltage
ripple.
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
D. Stage 4
In this stage, S1 and S2 turn on; the active clamp switch Sc3
is in on state. At t3, the primary current of L1 and L2 increases
while that of L3 decreases. The antiparallel diode of Sc3
conducts. In this interval, L1 and L3 are in forward mode and L2
in flyback mode. The secondary windings of L1 and L2 are
series connected to charge the output capacitor Co through D02
and D03 and these of L3 and L2 are series connected to charge
Co through D03 and D06. The secondary currents of L1 L3 and L2
are thus given by Eqs. 4,5 and 3, respectively.
V
N in  Vout
(4)
1
D
iLK 1  iLK 1(t3 ) 
(t  t3 )
LLK 1
iLK 3 
N
Vin
 Vout
1 D
(t  t3 )
LLK 3
(5)
State 4 [t3-t4]: At t3, S1 turns OFF and S3 turns ON, which
turns Do2 ON. The primary side of coupled inductor L1 charges
the battery by S3. During this state, L2 operates in the Forward
mode and L1 operates in the Flyback mode to transfer energy to
the load. When S1 turns ON and Do2 turns OFF, a new
switching period begins.
In mode 2, the battery transfers energy to the load, as shown
in Fig. 44(a). It indicates the nighttime operation of the
standalone system. The circuit works as the Flyback-Forward
converter, where S3 and S4 are the main switches, Cc, S1 and S2
form an active clamp circuit. When the load is disconnected,
the standalone system enters into mode 3. The PV array
charges battery without energy transferred to the load due to
the opposite series connected structure of coupled inductor, as
shown in Fig. 4(b). S1 and S2 work simultaneously and the
topology is equivalent to two paralleled buck-boost converters.
E. Stage 5
The clamp switch Sc3 turns off at t4. Then Cc is resonant
with the leakage inductance of L3. The voltage across Sc3
increases from zero and ZVS is achieved on Sc3. S3 is turned on
when its parasitic diode is on and S3 is on with ZVS. At t4, S1
turns off, and then its parasitic capacitor is charged from the
magnetizing current of L1. In this period, the switching voltage
of S1 is clamped by Cc; L1 and L3 are in forward mode; and L2 is
in flyback mode. Furthermore, the second sides of L1 and L2 are
series connected to charge Co through D02 and D03 and these of
L2 and L3 are also series connected to charge Co through D03
and D06.
Vgs1
Vgs1
Vgs3
Vgs2
Vgs4
Vgs2
Vgs2
Vgs1
iL2a
iL1a
iL1a
iL2a
ib
iS1
vds1
vds1
iS1
vDo2
vDo2
iDo2
iDo2
t0
t2
t1
t4
t3
Fig. 3 Signal waveforms of the proposed converter under mode 1.
Cs1
S1
Cs3
Do
**************.
Cs2
S2
S4
L Lk
Do1
Co1
n2
+
+
-
(1)
(2)
State 3 [t2-t3]: At t2, S2 turns ON, which makes the two
coupled inductors work in the Flyback state to store energy
again, and Do2 is reverse-biased. The energy stored in Co1 and
Co2 transfers to the load. At t3, the leakage inductor current
decreases to zero and diode Do1 turns off with zero-current
switching operation.
S3
+
VB
n2
n1
*
Do2
•
*
Vo
-
Cs4
L1
L2
Co2
(a) Mode 2
Cs1
S1
Cs2
S2
Cs3
Do
S4
L Lk
Do1
Co1
n2
+
+
Cc
-
S3
n1
*
(a) [t0-t1]
-
n1
NVB
1 D
N
VB  N (VB ) 
D
D
Vpv
Vab
Cc
Vpv
According to the voltage balance law,
DVPV  (1  D)VB
Ro
L1
Ro
+
VB
Vo
-
Cs4
n1
•
L2
n2
*
Do2
Co2
(b) Mode 3
(b) [t1-t2]
Fig. 4. Converter operating modes.
(c) [t2-t3]
III. PERFORMANCE ANALYSIS AND FEEDBACK LOOP DESIGN
(d) [t3-t4]
Fig. 2 Four operating states of the proposed converter in mode 1.
In order to realize flexible energy flow, the modulation
strategy is proposed by combing PWM with PS. Firstly, the
iL1a
iL1a
iL2a
iL1a
iL2a
t
t
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
(a)
Case 1: 0<φ<φ crit1
1S1
S33
Case 3: φ crit2<φ<φ crit3
(c)
D
S1S1
1
S22
S44
t
Case 2: φ crit1<φ<φ crit2
(b)
D with duty ratio and FS needs to be
relationship of voltage gains
S
S
S
researched. In the following
analysis,
S1 andS S2 share the same
S
S
duty ratio, D, meanwhile, S3 and S4 are applied to the same duty
ratio. The gate signals
for S1 and S3 are complementary, as well
V
as for S2 and S4.
2
iL2a
S33
S
S2
φ
S1
S2 S2
SS
4 4
φ
NVB/D
NVB/D
ab
Vab
iLk
t
A. Analysis of Circuit Performance (D>0.5)
When the duty
cycle D>0.5, there are five operating cases
i
i
for analysis, as shown
in Fig. 5.
t
In case 1, the phase shift angle is between 0 and φcrit1. From
(d) inductor
Case 4: φ current,
<φ<φ
the waveform of the leakage
the secondary
side of the coupled inductor is equivalent to a discontinuous
conduction mode (DCM) of a Buck converter. When φ=φcrit1,
the current pulses A and B is in a boundary conduction mode.
iLk
t
L1a
iL1a
L2a
iL2a
crit3
D
S1S1
S4
φ
Fig. 5. Five operational cases for D > 0.5.
For pulse A, the current decreases to the minimum value and
increases
to zero during the time period of
(1-D)Ts, The
D
D
S
S
S
S
S
decrementStime is equal Sto Tscrit1 / 2 andSthe incrementS time
S1
1S1
S4
φ
S2
SS
44
(a) Case
Case 1:
1 0<φ<φ
(a)
L2a
S4
φ
S2
S2SS22
S
S133
S
SS22
crit2
S33
crit3
1
1 1
33
1
4 S4
2
2
4 4
2
NVB/D
NV
NV
B/D
B/D
Vab
crit2
S22
4
φ
NVB/D
crit1
D
S1S1
S1
SS
44
S4
S
4
t
2: φ
<φ<φ
Case 3: φ
<φ<φ
TheCase
secondary
side of the coupled inductor(c)is equivalent
to
D
S
S
two Buck converters
connected
in parallel SSat the DCM
SS
S
S
S
S
SS
S S ratio
S
operational
condition.
The
corresponding
equivalent duty
φ
φ
of the buck converter is φ/2π. According to NVthe/D equation of
voltage gain of Buck converter
in DCM, the output voltage can
V
be expressed as:
(b)
crit1
D
S33
S1S1S
t
L1a
iL2a
t
S1
NVB/D
Lk
iL1a
iL2a
S4 S4
ab
iLk
iL1a
S1
S33
S22
4
φ
NVB/D
t
SS
44
1S1
1
S
44
is (1  D  crit1 / 2 )Ts . Following the voltage-second balance
V
(3), the critical phase angle can be determined
in (4).
Vo

NVB crit1

 Ts  t (1i  D  crit1 )Ts
D
2
2
2
(3)
Vout
i
crit1    D  (1 i D) 
t
N  VB
(4)
Vab
iLk
S33
S
S33
S22
NVB/D
Vab
D
S1S1
(e) Case 5
Case 5: φ crit4<φ<2π
(e)
crit4
S33
S
S22
S
t
Vab
B
Vab
ab
iLk
iLk
iLk
t
t
iL1a
iL1a
t
t
: 0<φ<φ crit1
D
S33
S1 S
S4
S
4
Case
2: (d)
φ2
(b)
Case
S
S133
S
SS
44
S4
φ
D
S1S
S1
1
SS22
SS
44
iLk
S4 S4
S2
S4
2
φ
Vab
iLk
t
t
t
t
t
iL2a
(a)
φ
<φ<φ crit2
d)crit1 Case 4:
φ crit3<φ<φ crit4
i
S4
S
4
S2
SS
4 4
φ
S33
SS
1
S33
φ
crit2
S33
33
S22
1S1
11
2
2
S
44
S33
t
V
NV
B crit1
 o (1  D)
DLk 2
2 Lk (c) Case 3: φ
Case 2: φ crit1<φ<φ crit2
Case 5: φ crit4
<φ<2π
2
33
iLk
t
t
iL1a
Lk
L1a
t
(d) Case 4
t
crit2 <φ<φ crit3
1
2
4 4
B
(d)
Case 4: φ crit3<φ<φ crit4
Case 5: φ crit4<φ<2π
(7)
(8)
S
S
In case 3, the PS angle ranges from φcrit2
to φcrit3. SThe duty
SS
S S
ratio of the secondaryS sideφ equivalent
Buck converter
stays
constant at the value of 1-D, and the voltage
gain stands the
NV /D
V
highest, therefore,
the critical point, φcrit3, and the
corresponding voltage can be expressed as Eqs. 9 and 10. With
i
the increase in
PS angle, the voltage declines. φcrit3 is t the
boundary point
between
DCM and CCM. In this case, the
i
output voltagei cannot be controlled by PS angle, as shown in
t
Eq.11.
5: φ
<φ<2π
crit 3 (e)
 2Case
 crit
2
(9)
ab
(e)
t
iL2a
D
S1S1
NVB/D
t
S4 S4
NVB/D
NVB/D
L1a
Vab
iL2a
S1
S22
4
t
SS22
iL2a
t
crit4
crit3
t
(b)
(e)
iLk
t
iLk
NVB/D
Vab
NVB
D
t
V 
NV
t B
(1  D)  o crit 2
DLk i
2 Lk 2
S1
S2 S2

t
(c) Case 3
D
S1S1 S
4  2 Lk
1i 1 
2
R

T

o
s ( / 2 ) / 2
i
ab
Case 1:
(c)0<φ<φ
Casecrit1
3: φ crit2<φ<φ crit3
D
S1S1
S2
iL2aL2a
2
(5)
t
In case 2, the phase shift angle is between φcrit1 and φcrit2.
(e)
Case 5: φ
<φ<2π
Case 3: φ
<φ<φ
The(c) phase
shift angle
φcrit2 is the boundary point from
D DCM, which can be
continuous
conduction
mode (CCM) to
SS
SS
S
S
S
S
S
S
S
determined
by
(6).
The
voltage
equation
at φcrit1 and
φcrit2 can
SS
S S
φ
be expressed as (7) and (8).
NV /D
V
4  (1  D) N  VB
crit 2 

D
Vout
(6)
iL1a
iL1a
iL1a
iL2a
iLk
t
L2a
B
Vab
iLk
iL1a
iL1a
φ
Vab
t
t
4 4
NVB/D
NVB/D
NV
B/D
NV
B/D
Vab
SS22
DD
S1S
S11S1
S1
S33
S
S1
S33
S
S4
φ
S22
S
Case 4: φ crit3<φ<φ crit4
crit1 <φ<φ crit2
(b)
D
S1S1
iL2a
iLk
L1a
iL1a
iL2a
iL2a
Vo  2 
t
L2a
crit4
2
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
D
SS1
SS33
SS22
φ
V
D
NVB / D
(1S1 D)  o (1  SS1crit 3 )
Lk
2 Lk
2
iLk
S4
φ
S2
SS44
2
Vo  2 

4  2 Lk
Vab
Ro  Ts  (1
 D) 2 / 2
1 1
NVB/D
Vab
SS33
1
(10)
SS22
φ
NVB
D
1
3
2
4
B
B
iLk
Vo  2 
4  2 Lk
Ro  Ts  (1   / 2 )D2 / 2
(14)
SS2
B. Analysis of Circuit Performance (D<0.5) SS4
φ
Similarly, there are five operating cases when the duty cycle
D<0.5. The respective waveforms are shown in Fig. 6.
2
4
Vab
D
SS1
SS33
SS22
φ
iLk
tt
S4
φ
S2
(b) case 2: φcrit1
(c)(a)
case
3: φ
<φ<φ
crit2
crit3
case
1:
0<φ<φ
(c)
Case
3 crit1
DD
SSS1
SS11
S42
S
SSS
S3333
S1
S4
φ
SS44
φ
SS22
D
D
S
SSS111
S4
φ
SS1
1
SS44
SS22
NVB/D
NVB/D
S4
SS22
SS44
Lk
iiLk
t
t
(b)
2:4:φφcrit1
<φ<φcrit2
(d)case
case
crit3<φ<φ
(d)
Case
4 crit4
D
SS11
φ
SS22
SS44
(e) case
φ
(c) case
3: φ5:
crit2
S1
SS33
S4
S4
NVB/D
NVB/D
Vab
SS33
D
SS1
S1
SS22
SS44
SS33
SS22
S4
φ
iLk
S1
SS4 4
t
t
NVB/D
NVB/D
(e) (e)
caseCase
5:Vφcrit4
5 <φ<2π
ab <φ<φcrit4
(d) case 4: φVcrit3
Vab
φ
Vab
ab
iLk
iLk
S1
S3
SSS
3
NVB/D
Vab
Vab
tt
D
SS1
S1
SS44
iLk
NVB/D
Vab
crit1
SS33
SS33
SS22
S4
φ
S2
iLkiLk
t
NVB

Dcase
(e)
φcrit4<φ<2π
(a) case
1: 5:
0<φ<φ
SS1
SS4 4
SS44
NVB/D
In case 5 (φcrit4 <φ<2π), the duty ratio of secondary side
equivalent Buck converter is 1-φ/2π. The output voltage can be
expressed as
iiLk
Lk
(d) case 4: φcrit3<φ<φ
1 crit4 1 
SS22
S4 SS22
φ
D
SS1
SS11
VV
abab
VVabab
t2
SSS3S333
NVB/D
3
2
4
Vab
SS44
(11)
In case 4, the PS angle ranges from φcrit3 to φcrit4. The
leakage inductor current is still higher
than zero before next
iLk
voltage pulse. φcrit4 is the boundary
point between DCM and
t
CCM. The voltage gain stands as the highest,
therefore,
thecrit2
(b) case
2: φcrit1 <φ<φ
case 1: 0<φ<φcrit1
critical(a) point,
the critical point φcrit4 and the corresponding
D
D
voltage
can be expressed as Eqs. 12 and 13.SSSD1
SS
S1
SS1
S11
SS333
SS3


2



SS2 SS4
SS4 SS22
crit 4 S4
crit1
SS2
SS4
(12)
4
φ
φ
φ
crit 4
Vo
NV
(1 
)
(1  D)
NV /D
NV 2
/D
DLk

2 Lk
(13)
1
DD
SSSS11
S1
NVB/D
ab
Fig. 6. Five operating cases under D < 0.5.
iLk
iLk
t
(a) case
0<φ<φ
(a) 1:
Case
1 crit1
SS33
DD
SSSS11
S1
SS44
S4
φ
S2
SS3SS
333
SS44
SS22
S1S1
SS22
φ
SS44
NVB/D
NVB/D
1
S4
NVB/D
t
(b) case(b)
2: φcrit1 <φ<φ
2 crit2crit4
(d) case 4:Case
φcrit3<φ<φ
0<φ<φcrit1
SS33
D
SS11
S1
SS22
crit3<φ<φcrit4
S4
S1
SS33
φ
SS44
SS22
S4
NVB/D
NVB/D
Vab
iLk
t
t
(e) case 5: φcrit4<φ<2π
4
B
VabVab
iLk
iLk
3
2
Vab
Vab
t
For case 1 (0<φ<φcrit1), considering
the waveform of the
iLk
leakage inductor current, t the secondary side of coupled
inductor
equivalent
to a DCM Buck converter, the
(b)
case 2: φcrit1is<φ<φ
crit2
(c) case 3: φcrit2<φ<φcrit3
corresponding duty ratio is D=φ/2π.
D
D
Vo
2
S
SS1 SS11
SS3 SS33
1 S1    D 
crit1
N
 VB
S4
SS22 SS4
SS44 SS2
S4
(15)
φ
φ
NVB
2
Vo  2 NV /D NV /D

B
D
4  2 Lk
1 1
2
Ro  Ts  ( / 2 ) / 2
(16)
In case 2 (φcrit1<φ<φcrit2), the current of leakage inductor is
still above zero beforet next
pulse coming. The system
t
equations can be expressed as
(c) case(e)3:case
φcrit2<φ<φ
5: φcrit4crit3
<φ<2π
N  VB
crit 2  4 
Vo
(17)
Vo
NVB crit1


D
DLk 2
2 Lk
(18)
iLkiLk
t
V 
NVB
 D  o crit 2
DLk
2 Lk 2
(19)
In case 3 (φcrit2<φ<φcrit3), the duty ratio of secondary
winding equals to D. The system equations are
t
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
crit 3  2  crit 2
(20)
V

NVB
 D  o (1  crit 3 )
DLk
2 Lk
2
2
Vo  2 
1 1
4  2 Lk
Ro  Ts  D 2 / 2
Vpv
(21)
Vo  2 
MPPT loop
Vo
In case 5 (φcrit4<φ<2π), the duty ratio of secondary winding
can be expressed as 1-φ/2π. The output voltage is
2
4  2 Lk
1 1
Ro  Ts  (1   / 2 ) 2 / 2

NVB
D
(25)
From the above analysis, the duty cycle D is the control
variable to balance the PV voltage and battery voltage, and φ is
employed to control secondary output voltage. The twofreedom control approach makes the solar energy, battery and
load fully controllable. One condition should be applied to
achieve decoupled control performance, which can be
expressed as 0<φ<φcrit2 and φcrit3 <φ<2π. If the condition is not
satisfied, the secondary output voltage is determined by the
switching duty cycle rather than the phase shift angle, as
presented in Eqs. 11 and 22. In mode 1, the primary side is
equivalent to an interleaved Buck-Boost, which operates in the
continuous conduction mode due to the asymmetrical
complementary operation of the switching devices (S1, S3) and
(S2, S4), and the operation of the second side follows Buck
converter, where the corresponding duty ratio can be controlled
by phase shift angle except for case 3.
C. Feedback Loop Design (D>0.5)
In mode 1, S1 and S3 complementarily conduct, and the
on/off operation of S2 and S4 is complementary. When the
output energy of PV array is lower than the load power, the
battery should supply the required energy to sustain output
power. The primary side of the proposed converter can be
equivalent to a bidirectional buck-boost converter, while the
secondary side can be equivalent to a Buck converter in
discontinuous conduction mode. The maximum power point
tracking (MPPT) can be implemented by adjusting duty cycle
of switching devices. The output voltage can be controlled by
PS of the primary side bridge arm, which can be approximated
to adjust the duty cycle of Buck converter of secondary side to
realize output voltage regulation. The control block diagram of
the proposed three-port converter for PV-battery hybrid power
generation system is illustrated in Fig. 7.
C
S1
S3
C
S2
S4
0.8· Vpv_open
NVB

D
(22)
In case 4 (φcrit3<φ<φcrit4), the current of leakage inductor is
still above zero. The system equations can be expressed as
crit 4  2  crit1
(23)
crit 4
Vo
NVB
 (1 
)
D
DLk
2
2 Lk
(24)
PI
Phase
Shift
PI
Vo_ref
φ
Output voltage control loop
Fig. 7. Diagram of the proposed control scheme.
In the MPPT loop, the PV voltage can be regulated to
follow the optimal operating point, which is initially assigned
to 80% of the open circuit voltage of the PV array. The optimal
operating point can be determined by the outer MPP Tracker,
which has been broadly investigated in prior works [34]. The
PV voltage regulation loop is an effective feature to improved
MPPT performance as claimed in [35]. In the output voltage
control loop in Fig. 7, the phase angle is the control variable.
The phase angle of the carrier waveform for modulation is
shifted to realize output voltage loop control. The mode can be
switched smoothly by the proposed control method because of
the block diode of PV input, mode 1 can be transferred to mode
2 by changing D>0.5 to D<0.5, furthermore, the mode can be
switched from 1 to 3 by controlling the phase shift angle
between S1 and S2.
IV. SIMULATION AND EXPERIMENTAL RESULTS
Both simulation and experimental test are conducted to
evaluation the proposed system and control scheme. The
system parameters for evaluation are listed in Table I.
TABLE I
PARAMETERS OF THE PROTOTYPE CONVERTER
Parameter
Do1-Do2
S1-S4
N = n1:n2
Cc
Switching Frequency
Battery voltage
Load resistance
Co1-Co2
Step-up ratio
Product/Value
BYW99W200
FDP047AN
1:2
100V/100μF
20KHz
12V
90Ω
250V/470μF
6.25
A. Simulation Results
Simulation is carried out in PSIM environment to establish
the relationship of phase shift angle and output voltage, and to
testify the proposed control strategy in MPPT and output
voltage control.
Fig. presents the simulation results indicating the phase
shift angle control and output voltage response at D>0.5 and
D<0.5. The voltage gain can be divided into five cases which
coincide with the theoretical analysis. In case 1, 2, 4, and 5, the
output voltage is controllable by the phase shift while in case 3,
the output voltage cannot be controlled by phase angle, as in
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
Eqs. 11 and 22. When D<0.5, the relation becomes more linear
than that for D>0.5.
Case 1 Case 2 Case 3 Case 4 Case 5
80
70
Vo(V)
60
50
40
30
operation enters mode 2. In mode 3 (Fig. 13), S1 and S2 are
applied with the same duty ratio without a phase shift angle.
Therefore, the energy is transferred only from the PV array to
the battery. The output voltage becomes zero due to the reverse
series connection of the second side of the coupled inductor.
Fig. 14 illustrates the measured converter efficiency. It can be
seen that the maximum efficiency of this converter is 91.3% at
200W and the rated efficiency is approximately 90%.
20
0
50
100
150
200
250
300
350
400
iin
10A/div
Phase shift angle (°)
(a) D > 0.5
Case 1 Case 2 Case 3
60
Case 4 Case 5
iLK
2A/div
50
Vo(V)
40
30
iB
10A/div
20
10
(a) Case 1 (φ=30°)
0
0
50
100
150
200
250
300
350
400
iin
10A/div
Phase shift angle (°)
(b) D < 0.5
Fig. 8. Simulation results for phase shift angle control.
B. Experimental Results
The proposed inverter topology and control scheme are
implemented in a 250W prototype (see Fig. 9) with a Texas
Instruments TMS320F28335 controller.
Experimental tests are conducted with a PV array simulator
(Agilent Technology E4360A) to obtain the steady-state
waveforms of the proposed converter in different operating
cases. Fig. 10 illustrates waveforms of the input current (iin),
battery current (iB) and the second side of the coupled inductor
current (iLK), for the phase angle shift under different cases.
Fig. 11 shows the regulation performance of the converter.
Using the phase shift modulation, the output voltage is
controlled at the constant 80V level. Meanwhile, the duty cycle
control of PWM regulates the PV voltage at 12.8V, which
represents the operating point of MPP of the PV array
simulator.
Controller
iLK
2A/div
iB
10A/div
(b) Case 2 (φ=120°)
iin
10A/div
iLK
2A/div
iB
10A/div
(c) Case 3 (φ=180°)
iin
10A/div
iLK
2A/div
Load
iB
10A/div
Battery
(d) Case 4 (φ= 240°)
Fig. 10. Current waveforms for different cases (D >0.5 mode 1).
Converter
Uout
(50V/div)
Source
iLK
(5A/div)
Fig. 9. Experimental setup of the proposed converter test system.
Modes 2-3 are also tested with the results presented in Figs.
12 and 13. Fig. 12 shows the current waveforms with the phase
shift for different cases when D<0.5. Under this condition, the
battery discharges energy to the load and the blocking diode
prevents the reverse current into the PV array. The converter
UPV
(10V/div)
20μS/div
(a) Output voltage
100
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
95
Efficiency (%)
Uout
(50V/div)
iB
(5A/div)
UPV
(10V/div)
85
80
100
150
200
250
Power (W)
20μS/div
(b) PV voltage
90
Fig. 14. Experimental results for the converter efficiency.
Fig. 11. Experiment results of voltage regulation performance.
step-up capability interfacing the PV array, the battery storage,
and the isolated consumption. Three operating modes are
V. CONCLUSIONS
analyzed and have shown the effective operation of the
This paper has presented an isolated three-port DC-DC proposed topology for PV applications.
converter for standalone PV systems, based on an improved
From simulation and experimental test results, it can be seen
Flyback-Forward topology. The converter can provide a high that the output voltage and the PV voltage can be controlled
i (10A/div)
i (10A/div)
i (10A/div) by the phase shift and PWM respectively.
independently
The
decouple control approach is a simple and effective way to
i (2A/div)
achieve the regulation of output voltage and PV voltage, which
is important for the MPPT operation of standalone PV power
i (2A/div)
i (2A/div)
systems. In addition to developing simulation
models, a 250W
i (10A/div)
i (10A/div)
i (10A/div)
converter is prototyped and tested to verify the effectiveness
of
the proposed converter topology and control scheme.
(a) case(a)
1: φCase
= 30°1
(b) case 2:
φ = 120°
(c) case 3:
180°
The
developed technology is capable
ofφ =achieving
MPPT,
high conversion ratio and multiple operating modes whist
(10A/div)
i (10A/div)
i (10A/div)
making
the converter relatively relative simple, light, efficient
and cost-effective.
in
in
in
LK
LK
LK
B
B
iin
B
in
in
iLK (2A/div)
REFERENCES
iLK (2A/div)
iB (10A/div)
case 1: φ = 30°
iLK (2A/div)
iB (10A/div)
(b) case 2: φ = 120°
(b) Case 2
iin (10A/div)
iLK (2A/div)
iin (10A/div)
iLK (2A/div)
iB (10A/div)
e 2: φ = 120°
iB (10A/div)
(c) case
φ = 180°
(c) 3:Case
3
Fig. 12. Current waveforms for different cases (D<0.5 mode 2).
VGS1
5V/div
VGS2
5V/div
Vo
5V/div
iB
5A/div
Fig. 13. Steady-state waveforms under mode 3.
[1] F. Zhang, K. Thanapalan, A. Procter, S. Carr, J. Maddy, "Adaptive Hybrid
iB (10A/div)
Maximum
Power Point Tracking Method for a Photovoltaic System,”
IEEE Trans. Energy Conversion, vol. 28, no. 2, pp. 353-360, June 2013.
Thounthong, “Model Based-Energy Control of a Solar Power Plant
(c) case[2]
3: φ P.
= 180°
With a Supercapacitor for Grid-Independent Applications,” IEEE Trans.
Energy Conversion, vol. 26, no. 4, pp. 1210-1218, 2011.
[3] A. Elmitwally, M. Rashed, “Flexible Operation Strategy for an Isolated
PV-Diesel Microgrid Without Energy Storage,” IEEE Trans. Energy
Conversion, vol. 26, no. 1, pp. 235-244, 2011.
[4] J. K. Shiau, D. M. Ma, P. Y. Yang, G. F. Wang, and J. H. Gong, “Design
of a Solar Power Management System for an Experimental UAV,” IEEE
Trans. Aerospace and Electronic Systems, vol. 45, no. 4, pp. 1350-1360,
Oct. 2009.
[5] F. S. Kang, S. J. Park, S. E. Cho, C. U. Kim, T. Ise, "Multilevel PWM
inverters suitable for the use of stand-alone photovoltaic power systems,”
IEEE Trans. Energy Conversion, vol. 20, no. 4 , pp. 906-915, 2005.
[6] H. Valderrama-Blavi, J. M. Bosque, F. Guinjoan, L. Marroyo, and L.
Martinez-Salamero, “Power Adaptor Device for Domestic DC Microgrids
Based on Commercial MPPT Inverters,” IEEE Trans. Ind. Electron., vol.
60, no. 3, pp. 1191-1203, Mar. 2013.
[7] H. Wu, K. Sun, R. Chen, H. Hu, and Y. Xing, “Full-Bridge Three-Port
Converters With Wide Input Voltage Range for Renewable Power
Systems,” IEEE Trans. Power Electron., vol. 27, no. 9, pp. 3965-3974,
Sept. 2012.
[8] O. Elma, and U. S. Selamogullari, “A comparative sizing analysis of a
renewable energy supplied stand-alone house considering both demand
side and source side dynamics,” Applied Energy, vol. 96, pp. 400-408,
2012.
[9] H. Wu, R. Chen, J. Zhang, Y. Xing, H. Hu, and H. Ge, “A Family of
Three-Port Half-Bridge Converters for a Stand-Alone Renewable Power
System,” IEEE Trans. Power Electron., vol. 26, no. 9, pp. 2697-2706,
Sept. 2011.
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
[10] Z. Qian, O. Abdel-Rahman, and I. Batarseh, “An Integrated Four-Port
DC/DC Converter for Renewable Energy Applications,” IEEE Trans.
Power Electron., vol. 25, no. 7, pp. 1877-1887, July 2010.
[11] E. Sanchis, E. Maset, A. Ferreres, J. B. Ejea, V. Esteve, J. Jordan, J.
Calvente, A. Garrigos, and J. M. Blanes, “Bidirectional High-Efficiency
Nonisolated Step-Up Battery Regulator,” IEEE Trans. Aerospace and
Electronic Systems, vol. 47, no. 3, pp. 2230-2239, Jul. 2011.
[12] K. Shenai, P. G. Neudeck, and G. Schwarze, “Design and technology of
compact high-power converters,” IEEE Aerospace and Electronic Systems
Magazine, vol. 16, no. 3, pp. 27-31, Aug. 2001.
[13] S. Kim, H.-S. Song, and K. Nam, “Idling Port Isolation Control of ThreePort Bidirectional Converter for EVs,” IEEE Trans. Power Electron., vol.
27, no. 5, pp. 2495-2506, May 2012.
[14] C. Zhao, S. D. Round, and J. W. Kolar, “An Isolated Three-Port
Bidirectional DC-DC Converter With Decoupled Power Flow
Management,” IEEE Trans.Power Electrons., vol. 23, no. 5, pp. 24432453, Sept. 2008.
[15] H. Tao, J. L. Duarte, and M. A. M. Hendrix, “Three-Port Triple-HalfBridge Bidirectional Converter With Zero-Voltage Switching,” IEEE
Trans. Power Electron., vol. 23, no. 2, pp. 782-792, Mar. 2008.
[16] J. L. Duarte, M. Hendrix, and M. G. Simoes, “Three-Port Bidirectional
Converter for Hybrid Fuel Cell Systems,” IEEE Trans. Power Electron.,
vol. 22, no. 2, pp. 480-487, Mar. 2007.
[17] H. Wu, P. Xu, H. Hu, Z. Zhou, and Y. Xing, “Multi-port Converters
Based on Integration of Full-Bridge and Bidirectional DC-DC Topologies
for Renewable Generation Systems,” IEEE Trans. Ind. Electron., vol. PP,
no. 99, pp. 1-1, in press, 2013.
[18] Z. Ouyang, Z. Zhang, M. A. E. Andersen, and O. C. Thomsen, “Four
Quadrants Integrated Transformers for Dual-Input Isolated DC-DC
Converters,” IEEE Trans. Power Electron., vol. 27, no. 6, pp. 2697-2702,
June 2012.
[19] L. Wang, Z. Wang, and H. Li, “Asymmetrical Duty Cycle Control and
Decoupled Power Flow Design of a Three-port Bidirectional DC-DC
Converter for Fuel Cell Vehicle Application,” IEEE Trans. Power
Electron., vol. 27, no. 2, pp. 891-904, Feb. 2012.
[20] R. J. Wai, C. Y. Lin, and Y. R. Chang, “High Step-Up Bidirectional
Isolated Converter With Two Input Power Sources,” IEEE Trans. Ind.
Electron., vol. 56, no. 7, pp. 2629-2643, July 2009.
[21] W. Li, L. Fan, Y. Zhao, X. He, D. Xu, and B. Wu, “High-Step-Up and
High-Efficiency Fuel-Cell Power-Generation System With Active-Clamp
Flyback-Forward Converter,” IEEE Trans. Ind. Electron., vol. 59, no. 1,
pp. 599-610, 2012.
[22] Z. Qian, O. Abdel-Rahman, H. Al-Atrash, and I. Batarseh, “Modeling and
Control of Three-Port DC/DC Converter Interface for Satellite
Applications,” IEEE Trans. Power Electron., vol. 25, no. 3, pp. 637-649,
2010.
[23] P. Bajpai, and V. Dash, “Hybrid renewable energy systems for power
generation in stand-alone applications: A review,” Renewable and
Sustainable Energy Reviews, vol. 16, no. 5, pp. 2926-2939, 2012.
[24] H. Tao, A. Kotsopoulos, J. L. Duarte, and M. A. M. Hendrix, “Family of
multiport bidirectional DC-DC converters,” IEE Proc. Electric Power
Applications, vol. 153, no. 3, pp. 451-458, May 2006.
[25] B. G. Dobbs, and P. L. Chapman, “A multiple-input DC-DC converter
topology,” IEEE Power Electronics Letters, vol. 1, no. 1, pp. 6-9, Mar.
2003.
[26] Z. Wang, and H. Li, “An Integrated Three-Port Bidirectional DC-DC
Converter for PV Application on a DC Distribution System,” IEEE Trans.
Power Electron., vol. 28, no. 10, pp. 4612-4624, Oct. 2013.
[27] W. Cai, B. Liu, S. Duan, L. Jiang "Power flow control and optimization
of a three-port converter for photovoltaic-storage hybrid system,"
Energy Conversion Congress and Exposition (ECCE), pp. 4121-4128,
2012.
[28] Y. M. Chen, X. Yu, A. Q. Huang, "A new nonisolated three-port DC-DC
converter with high step-up/down ratio,” the 2012 IEEE
Energy Conversion Congress and Exposition (ECCE), pp. 1520-1526,
2012.
[29] Q. Wang, J. Zhang, X. Ruan, and K. Jin, “Isolated Single Primary
Winding Multiple-Input Converters,” IEEE Trans. Power Electron., vol.
26, no. 12, pp. 3435-3442, Dec. 2011.
[30] Y. Li, X. Ruan, D. Yang, F. Liu, and C. K. Tse, “Synthesis of MultipleInput DC/DC Converters,” IEEE Trans. Power Electron., vol. 25, no. 9,
pp. 2372-2385, Sept. 2010.
[31] H. Krishnaswami, and N. Mohan, “Three-Port Series-Resonant DC-DC
Converter to Interface Renewable Energy Sources With Bidirectional
Load and Energy Storage Ports,” IEEE Trans. Power Electron., vol. 24,
no. 10, pp. 2289-2297, Oct. 2009.
[32] H. Tao, A. Kotsopoulos, J. L. Duarte, and M. A. M. Hendrix,
“Transformer-Coupled Multiport ZVS Bidirectional DC-DC Converter
With Wide Input Range,” IEEE Trans. Power Electron., vol. 23, no. 2,
pp. 771-781, Mar. 2008.
[33] W. Li, J. Xiao, Y. Zhao, and X. He, “PWM Plus Phase Angle Shift
(PPAS) Control Scheme for Combined Multiport DC/DC Converters,”
IEEE Trans. Power Electron., vol. 27, no. 3, pp. 1479-1489, 2012.
[34] W. Xiao, A. Elnosh, V. Khadkikar, and H. Zeineldin, “Overview of
maximum power point tracking technologies for photovoltaic power
systems,” in Proc. Annual Conference on IEEE Industrial Electronics
Society, 2011, pp. 3900-3905.
[35] W. Xiao, W. G. Dunford, P. R. Palmer, and A. Capel, “Regulation of
Photovoltaic Voltage,” IEEE Trans. Industrial Electronics, vol. 54, no. 3,
pp. 1365-1374, Jun. 2007.
Download