energies

Article
Modelling and Control of Parallel-Connected
Transformerless Inverters for Large
Photovoltaic Farms
Marian Liberos *, Raúl González-Medina, Gabriel Garcerá ID
and Emilio Figueres
Grupo de Sistemas Electrónicos Industriales del Departamento de Ingeniería Electrónica,
Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain;
raugonme@upv.es (R.G.-M.); ggarcera@eln.upv.es (G.G.); efiguere@eln.upv.es (E.F.)
* Correspondence: malimas@upv.es; Tel.: +34-963-879-606

Received: 14 July 2017; Accepted: 14 August 2017; Published: 21 August 2017

Abstract: This paper presents a control structure for transformerless photovoltaic inverters connected
in parallel to manage photovoltaic fields in the MW range. Large photovoltaic farms are usually
divided into several photovoltaic fields, each one of them managed by a centralized high power
inverter. The current tendency to build up centralized inverters in the MW range is the use of
several transformerless inverters connected in parallel, a topology that provokes the appearance of
significant zero-sequence circulating currents among inverters. To eliminate this inconvenience, this
paper proposes a control structure that avoids the appearance of circulating currents by controlling
the zero-sequence component of the inverters. A second contribution of the paper is the development
of a model of n parallel-connected inverters. To validate the concept, the proposed control structure
has been applied to a photovoltaic field of 2 MW managed by four 500 kW photovoltaic inverters
connected in parallel.

Keywords: photovoltaic farms; parallel inverters; circulating current; modeling and control

1. Introduction
In recent years the demand for electricity has increased. In the future, the global electricity
consumption is expected to continue growing significantly [1]. Moreover, the prices of the conventional
sources that have a limited stock are increasing, so renewable energies such as the photovoltaic (PV)
solar cells are becoming a promising alternative. Photovoltaic energy has many advantages: it is
a clean energy, it is omnipresent and has lower operational and maintenance costs [2]. Nowadays,
large-scale PV plants that improve the profitability of the photovoltaic systems are proliferating [3].
In the context of photovoltaic fields, there are many studies dealing with the use of decentralized
inverters that are connected to only one string with their own Maximum Power Point Tracking
(MPPT) [4]. The main benefit of the use of decentralized inverters is the high number of MPPT inputs
of these systems. However, decentralized topologies, illustrated in Figure 1a, are less interesting
for high power applications because of their high cost. As an alternative, centralized inverters are
mostly used in high power applications, because they offer a good compromise between costs and
efficiency [5]. Figure 1b–f shows various PV system schemes based on centralized inverters.

Energies 2017, 10, 1242; doi:10.3390/en10081242 www.mdpi.com/journal/energies

Energies 2017, 10, 1242 2 of 25
Energies 2017, 10, 1242 2 of 25

...
...

(a) (b) (c)

... ... ...

(d) (e) (f)

Figure 1. Topologies of high power PV inverters: (a) decentralized inverters; (b) centralized inverter;
Figure 1. Topologies of high power PV inverters: (a) decentralized inverters; (b) centralized inverter;
(c) centralized inverter composed by n parallel modules and a multi-output transformer;
(c) centralized inverter composed by n parallel modules and a multi-output transformer; (d) centralized
(d) centralized inverter composed by n parallel modules and n transformers; (e) centralized inverter
inverter composed by n parallel modules and n transformers; (e) centralized inverter composed by n
composed by n modules and a single transformer; (f) centralized inverter composed by n
modules and a single
transformerless transformer; (f) centralized inverter composed by n transformerless modules.
modules.

FigureFigure 1b illustrates
1b illustrates a central
a central inverter
inverter thatconnects
that connectsthe the whole
whole PV PV field
field to
tothe
thedistribution
distribution grid
grid by
by means of a low frequency transformer. The main problem of using
means of a low frequency transformer. The main problem of using this scheme is the poor efficiency this scheme is the poor
efficiency in the low power range, when the radiation conditions are low and the entire system must
in the low power range, when the radiation conditions are low and the entire system must remain in
remain in operation. In addition, the transformer is expensive and bulky. To partially overcome these
operation. In addition, the transformer is expensive and bulky. To partially overcome these drawbacks,
drawbacks, Figure 1c–f shows several solutions that implement a centralized inverter by means of
Figure the1c–f shows
parallel several of
connection solutions
modules,that eachimplement
one of them a centralized inverterofbythe
managing a fraction means of theThe
full power. parallel
connection of modules, each one of them managing a fraction of the full power.
parallel operation of inverters allows modularity and also the inverter disconnection depending on The parallel operation
of inverters
the levelallows modularity
of the generated power.andThis alsotechnique
the inverter disconnection
is frequently depending
used to improve on the level
the efficiency in theof the
generated power.
low power Thisastechnique
range, is frequently
Figure 3 depicts and willused to improve
be discussed theInefficiency
later. Figure 1c,d inthe
theuse
lowofpower
a singlerange,
multiple-output
as Figure 3 depicts transformer
and will beand an individual
discussed later. transformer
In Figure 1c,d per inverter
the useisofproposed,
a single respectively.
multiple-output
Both solutions take advantage of the paralleling principle,
transformer and an individual transformer per inverter is proposed, respectively. but they are up to 40% moreBoth expensive
solutions
than having a single secondary as in Figure 1e [3]. Although
take advantage of the paralleling principle, but they are up to 40% more expensive than the use of a single transformer for all
having
modules is not as expensive as the two previously described solutions, it is as expensive and bulky
a single secondary as in Figure 1e [3]. Although the use of a single transformer for all modules is not as
as the one of Figure 1b. At this point, the question is whether galvanic isolation is mandatory in
expensive as the two previously described solutions, it is as expensive and bulky as the one of Figure 1b.
photovoltaic installations or if it can be avoided to reduce costs, volume and weight, as well as power
At this point,
losses the system.
of the question It is
is whether
true that galvanic
transformersisolation
allow is mandatory
filtering out theinDC photovoltaic
componentsinstallations
of the
or if generated
it can be avoided
currents, soto that
reduce
thesecosts, volume
components areand
not weight,
injected to asthe
well as powergrid,
distribution losses
as itofisthe system.
stated
It is true that transformers allow filtering out the DC components of
by standards and regulations about the grid connection of power inverters. However, for this the generated currents, so that
thesepurpose
components
the use areofnot injected tocan
transformers thebedistribution
replaced bygrid, any as it is stated
technique thatby standards
limits and regulations
the generated DC
aboutcomponents below a certain
the grid connection threshold
of power that would
inverters. depend
However, foronthis
thepurpose
local regulations
the use of of each country. can
transformers
This factby
be replaced opens
anythe way to the
technique thatuselimits
of transformerless
the generated inverters in PV installations,
DC components belowwhich are much
a certain threshold
less bulky and expensive and also avoid the power losses produced
that would depend on the local regulations of each country. This fact opens the way to the by transformers. To illustrate this
use of
fact, Figure 2 shows the efficiency of the Power Gate Plus 500 kW, a high power inverter
transformerless inverters in PV installations, which are much less bulky and expensive and also avoid
the power losses produced by transformers. To illustrate this fact, Figure 2 shows the efficiency of the
Power Gate Plus 500 kW, a high power inverter manufactured by Satcon (Shiyan, China). In this figure
the efficiency in the whole power range with (in red) and without transformer (in green) has been
Energies2017,
Energies 2017,10,
10,1242
1242 33of
of25
25

manufactured
manufactured
Energies by Satcon
by
2017, 10, 1242 Satcon (Shiyan,
(Shiyan, China).
China). In
In this
this figure
figure the
the efficiency
efficiency in
in the
the whole
whole power
power range
range3with
with
of 25
(inred)
(in red)and
andwithout
withouttransformer
transformer(in (ingreen)
green)has
hasbeen
beenrepresented
representedaccording
accordingto tothe
thedatasheet
datasheet[6].
[6].ItItcan
can
be noted that the efficiency of the transformerless version of the inverter is significantly
be noted that the efficiency of the transformerless version of the inverter is significantly greater in the greater in the
represented
whole power according
whole power range. range. to the datasheet [6]. It can be noted that the efficiency of the transformerless
version of the inverter is significantly greater in the whole power range.

98.00%
98.00%
97.00%
97.00%
Efficiency(%)
Efficiency(%)

96.00%
96.00%
95.00%
95.00%
94.00%
94.00%
93.00%
93.00%
00 5500 110000 115500 220000 225500 330000 335500 440000 445500 550000
Power (kw)
Power (kw)

Figure
Figure2.2. Efficiency of
2.Efficiency
Efficiency the Power
ofthe
the Power Gate
Gate Plus
Plus 500
500 kW
kW from
from Satcon
Satcon with
with (in
(in red)
red) and
and without
without transformer
transformer
Figure of Power Gate Plus 500 kW from Satcon with (in red) and without transformer
(in green).
(in green).
green).
(in

This work
This work focuses
focuses on on the
the parallel
parallel connection
connectionof of nn transformerless
transformerless inverters
transformerless inverters
invertersas as Figure
asFigure
Figure1f 1f shows.
1fshows.
shows.
As stated
As stated above,
above, this
this connection
connection has many benefits
has many benefitssuch such as as modularity,
modularity, better better profitability
profitability and and lower
lower
volume, weight
weight and
and costs.
costs. The
The efficiency
efficiency of a
of centralized
a centralized inverter
inverter
volume, weight and costs. The efficiency of a centralized inverter constituted by a single module that constituted
constituted by a
bysingle
a module
single module that
manages
that
managesmanages all the
all the power
all power
the power is poor
is poor
is poor when
when when thethe
the generated
generated
generated power
powerpower is is
is low.
low.
low. As
AsAsaaasolution
solution
solutionto to this
tothis problem,
thisproblem,
topologies composed of inverters connected in parallel are usually
topologies composed of inverters connected in parallel are usually employed [7–9]. The main benefit employed
employed [7–9].
[7–9]. The main benefit
of these systems is that each inverter can be connected as the generated
of these systems is that each inverter can be connected as the generated power increases. To illustrate power increases. To illustrate
this fact,
this fact, Figure illustratesthe
Figure 33 illustrates efficiencyconsidering
the efficiency considering aa single single inverter
inverter which
which manages
manages the the full power
full power
(in red), and the efficiency
efficiency of
of four
four parallel
parallel inverters
inverters managing each
(in red), and the efficiency of four parallel inverters managing each one a fraction of the total power one a fraction of the total power
(in blue).
(in blue). The improvement of
The improvement of the efficiency is
the efficiency
efficiency is evident
is evidentat
evident at the
the lowlow power
power range,
range, in in which
which the the inverters
inverters
work for
work for many
for many hours.
many hours. Note
hours. Note thatthat thethe inverters
inverters can can bebe sequentially
sequentially connected
connected to to achieve
achieve the the better
better
efficiency profile in the whole range of power,
power, as
as Figure
Figure 3
3 shows.
shows.
efficiency profile in the whole range of power, as Figure 3 shows. The data to build up the efficiency The data to build up the efficiency
profiles have
profiles
profiles havebeen
have been
been obtained
obtained
obtained from from
from[6]. It[6]. ItIt is
is worth
[6]. is worth
worth
to pointto to
out point out that
that there
point out that there are
are companies
there are that
companies
manufacture
companies that
that
manufacture
photovoltaic photovoltaic
inverters in inverters
the MW in
range. the MW
To range.
achieve To
this achieve
power
manufacture photovoltaic inverters in the MW range. To achieve this power range, several modules this
range, power
several range, several
modules of modules
around
of around
500
of around
kW are 500 500 kW
usuallykWconnected
are usually
are usually connected
in parallel,
connected inisparallel,
as itin parallel,
the case of asthe
as itit is
is
HEC theV1000
the case of
case of
(Powerthe HEC
the HEC V1000 Paterna,
Electronics,
V1000 (Power
(Power
Electronics,
Valencia)
Electronics, Paterna,
[8]Paterna,
and the VS980 Valencia)
Valencia)solar[8][8] and the
inverter
and the
(ABB, VS980
VS980 solarSwitzerland)
Zürich,
solar inverter (ABB,
inverter (ABB, Zürich,
[9]. Zürich,
The switches Switzerland) [9]. The
market currently
Switzerland) [9]. The
switches
offers a market
large currently
variety of offers
stacks a
that large
could variety
be usedof stacks
in thesethat could be
applications.
switches market currently offers a large variety of stacks that could be used in these applications. For used Forin these
instance,applications.
the company For
instance, the company Infineon (Neubiberg, Germany) TM Module
has the PrimePACK TM Module series up to
instance, the company Infineon (Neubiberg, Germany) has the PrimePACK Module series upare
Infineon (Neubiberg, Germany) has the PrimePACK series up to 1700
TM V-1400 A, which to
1700 V-1400
V-1400 A,
recommended
1700 A,for
which
which are topologies
2-level
are recommended
recommended [10].for for 2-level
2-level topologies
topologies [10]. [10].

Inv22activation
Inv activation
Inv33activation
Inv activation
Inv44activation
Inv activation
98.00%
98.00%
Efficiency(%)

97.50%
Efficiency(%)

97.50%

97.00%
97.00% Centralizedinverter
Centralized inverter

96.50%
96.50%

96.00%
96.00%
00 220000 440000 660000 880000 11000000 11220000 11440000 11660000 11880000 22000000
Power (kw)
Power (kw)

Figure 3.
Figure 3. Efficiency of
of a singleinverter
inverter (red)vsvs several parallel-connected
parallel-connected inverters (blue).
(blue).
Figure 3.Efficiency
Efficiency of aa single
single inverter (red)
(red) vs several
several parallel-connected inverters
inverters (blue).
Energies 2017, 10, 1242 4 of 25
Energies 2017, 10, 1242 4 of 25

In spite of their advantages, the main problem of connecting parallel inverters is the appearance
of circulating
In spite ofcurrents that have the
their advantages, a zero-sequence
main problem nature. In parallel
of connecting operation,
parallel invertersthe converters
is the appearanceare
usually designed
of circulating individually,
currents that have butawhen they operate
zero-sequence in parallel,
nature. a zero-sequence
In parallel operation, circulating current
the converters are
that can produce malfunctions in the system appears [11]. To overcome this
usually designed individually, but when they operate in parallel, a zero-sequence circulating current problem, in [12–14], it
was proposed
that can produce themalfunctions
use of isolating
in thetransformers
system appears as in Figure
[11]. 1c,d, butthis
To overcome therefore
problem, aninexpensive
[12–14], it and
was
bulky
proposedsystem results.
the use In othertransformers
of isolating works [15,16], the
as in n parallel
Figure modules
1c,d, but are an
therefore treated as only
expensive andone by
bulky
considering an equivalent single converter that has as many legs as the number
system results. In other works [15,16], the n parallel modules are treated as only one by considering anof legs of each module
multiplied
equivalent by n. Nevertheless,
single converter thatthis hasapproach
as many doesn’t allow
legs as the modular
number design
of legs andmodule
of each has a great degree by
multiplied of
complexity, so that it is only suitable for low values of n.
n. Nevertheless, this approach doesn’t allow modular design and has a great degree of complexity, so
Inisthis
that it onlypaper
suitable a for
control technique
low values of n. that eliminates the problem of circulating currents by
controlling
In this the
paperzero-sequence components
a control technique of the generated
that eliminates currents
the problem is proposed.
of circulating To achieve
currents this, the
by controlling
proposed technique implements a three-dimension space vector modulator
the zero-sequence components of the generated currents is proposed. To achieve this, the proposed into the control loop of
ntechnique
− 1 inverters. The complexity
implements of the proposed
a three-dimension solution
space vector is veryinto
modulator lowtheand, as it loop
control will of
benshown in the
− 1 inverters.
next sections, it of
The complexity is compatible
the proposed with modular
solution design
is very low and
and,can
as itbe easily
will extended
be shown to next
in the high sections,
values ofitn.is
Another contribution of the paper is the development of an accurate model
compatible with modular design and can be easily extended to high values of n. Another contribution of these kinds of PV
systems.
of the paper is the development of an accurate model of these kinds of PV systems.

2.
2. Modeling of nn Parallel
Modeling of Parallel pv
pv Inverters
Inverters in
in aa Synchronous
Synchronous Reference
Reference Frame
Frame
Figure 44 depicts
Figure depicts the scheme of n PV transformerless inverters connected in parallel. The The grid
grid
filter for
filter for each
each inverter
inverter is is an LCL network that has damping resistors, Rdd,, connected
connected in in series
series to
to the
the
filter capacitors.
filter capacitors. InIn the
the model system the mutual coupling terms, Maa and
model of the system Mbb,,of
and M ofthe
thethree-phase
three-phase
inductors, and LLbb, ,have
inductors, Laa and havebeen
beentaken
takeninto
intoaccount.
account.ItItisisworth
worthtotopoint
pointout
outthat
thatthe
thecoupling
coupling terms
terms are
are
usually neglected,
usually neglected, even
even ifif three-phase
three-phase inductors
inductors instead
instead of ofone
oneinductor
inductorper
perphase
phaseare
areused.
used.However,
However,
they affect
they affect to
to the
the dynamic
dynamic response
response of
of the
the system,
system, so that it is interesting to consider them. In In the
the
model, the
model, the grid inductance LLgg has
grid inductance has been
been also
also considered.
considered.

i1a1 La Lb i Lg
+ 2a1 Vga
Vpv i1b1 i2b1 Vgb
Co
- i1c1 i2c1 Vgc vN

Rd
Cf
v’N1
...

...

...

i1an La Lb i
2an
i1bn i2bn
Co
i1cn i2cn
Rd
Cf
v’Nn
Figure
Figure 4.
4. Scheme
Schemeof
ofnnPV
PVtransformerless
transformerless inverters
inverters connected
connected in
in parallel
parallel with
with LCL
LCL grid
grid filters.
filters.

2.1.
2.1. Equations
Equations of
of the
the Averaged
Averaged Model
Model
Figure
Figure 5c
5cshows
showsthetheequivalent
equivalentcircuit
circuitofofonly
onlyone
oneinverter leg,
inverter which
leg, whichis derived byby
is derived averaging in
averaging
ainswitching cycle
a switching thethe
cycle grid voltage
grid and
voltage andthe
theDC DCcurrent
currentthat
thatappear
appearininFigure
Figure5a.
5a. In
InFigure
Figure 5b
5b the
the
Energies 2017, 10, 1242 5 of 25
Energies 2017, 10, 1242 5 of 25
instantaneous and averaged values of the variables are represented, being dφ, (φ = a, b, and c) the duty
Energies 2017, 10, 1242 5 of 25
cycle of the upper Insulated Gate Bipolar Transistor (IGBT) (i.e., of the control signal Sφp).
instantaneous and averaged values of the variables are represented, being dφ , (φ = a, b, and c) the duty
instantaneous and averaged values of the variables are represented, being dφ, (φ = a, b, and c) the duty
cyclecycle
of the upper Insulated Gate Bipolar Transistor (IGBT) (i.e., of the control signal S ).
of the upper Insulated Gate Bipolar Transistor (IGBT) (i.e., of the control signal Sφp). φp
Sφp

Sφp
Vpv vφ
Vpv vφ
Sφn
Sφn
(a)
dφT (a)
dφT
Sφp
Sφp
Sφn
Sφn
T T
vφ vφ
v̄ pv v̄ pv
ddφīφφīφ
+ +
iDC iDC v̄ vpv
̄ pv - -
dφdv̄ φpvv̄ pv
īφ īφ

(b) (c)
(b) (c)
Figure 5. Averaged equivalent circuit of an inverter leg: (a) Inverter leg; (b) Switching time and
Figure 5.
Figure Averaged equivalent
5. Averaged equivalent circuit
circuit of
of an
an inverter
inverter leg:
leg: (a)
(a) Inverter
Inverter leg;
leg; (b)
(b) Switching
Switching time
time and
and
relationship among input and output variables; (c) Averaged equivalent circuit.
relationship among
relationship among input
input and
and output
output variables;
variables; (c)
(c) Averaged
Averaged equivalent
equivalent circuit.
circuit.
By applying the averaged equivalent circuit to each one of the inverter legs in Figure 4, the
By
By applying
averaged
applyingmodelthe
theofaveraged equivalent
the n parallel
averaged circuit
transformerless
equivalent circuitPVto each
each one
toinvertersonein of
of the
the inverter
stationary
the inverter legs
legs in
in Figure
three-phase frame is4,
Figure 4, the
the
averaged model
obtained,
averaged model as of the
the nn6parallel
Figure
of shows. transformerless
parallel Equations (1)–(8)PV
transformerless PVareinverters in
in the
the stationary
directly obtained
inverters from thisthree-phase
stationary scheme. In frame
three-phase the
frame is
is
equations,
obtained, as i = 1,
Figure …,
6 n identifies
shows. each one
Equations of the modules
(1)–(8) are and n
directly is the total
obtained number
from
obtained, as Figure 6 shows. Equations (1)–(8) are directly obtained from this scheme. In the equations, of modules.
this scheme.NoteIn the
that Equations
equations, 1, …,(5)–(7) are equal for one
eachof
inverter if the components are total
considered of the
of same valueNote
i = 1, . . . , in=identifies
n identifies
each oneeach the modules
of the modules and n is andthen total
is the number number
of modules.modules.
Note that
and if tolerances around the nominal values are neglected.
that
Equations (5)–(7) are equal for each inverter if the components are considered of the same value value
Equations (5)–(7) are equal for each inverter if the components are considered of the same and if
and if tolerances
tolerances around around the nominal
the nominal valuesvalues are neglected.
are neglected.
ī1a1 La Lb ī Lg
2a1 Vga
ī1b1 ī2b1 Vgb
īī1c1
1a1
La Lbī2c1ī Lg Vgc vNVga
2a1
Co da1ī1a1 db1ī1b1 dc1ī1c1 da1v̄ pv d̄1v̄ pv
īpv ī1b1 Rd ī2b1 Vgb
+ + +
dc1v̄ pv
ī1c1 ī2c1 Vgc vN
- - -
Cf
Co da1ī1a1 db1ī1b1 dc1ī1c1 da1v̄ pv d̄1v̄ pv
īpv Rdv’N1
+ + +
dc1v̄ pv
...

...

...

- - -
Cf
ī1an La v’N1 Lb ī
2an
...

...

...

ī1bn ī2bn
ī1cn ī2cn
Co danī1an dbnī1bn dcnī1cn danv̄ pv dbnv̄ pv La Lb ī
ī1an Rd 2an
+
-
+
-
+
- dīcnv̄ pv ī2bn
1bn Cf
ī1cn v’Nn ī2cn
Co danī1an dbnī1bn dcnī1cn danv̄ pv dbnv̄ pv
Rd
Figure 6. Averaged equivalent circuit
- of
+
dcnv̄ pv +
- n PV transformerless
-
+
inverters connected in parallel.
Cf
v’Nn

Figure6.
Figure Averaged equivalent
6.Averaged equivalent circuit of nn PV
circuit of PV transformerless
transformerless inverters connected in parallel.
Energies 2017, 10, 1242 6 of 25

d d d
VLa i = La · i + Ma · i1bi + Ma · i1ci (1)
dt 1ai dt dt
d d d
VLb i = Ma · i1ai + L a · i1bi + Ma · i1ci (2)
dt dt dt
d d d
VLc i = Ma · i1ai + Ma · i1bi + L a · i1ci (3)
dt dt dt
i gϕ = i2ϕ1 + i2ϕ2 + . . . + i2ϕn = ni2ϕ (4)
             
L a Ma Ma i1ai d ai vcai i1ai i2ai vN0
 d
 Ma L a Ma ·  i1bi  = v pv  dbi  −  vcbi  − Rd  i1bi  + Rd  i2bi  −  v N 0  (5)
           
dt
Ma Ma L a i1ci dci vcci i1ci i2ci vN0
" # " # " # " # " # " # " # " #
Lb + nL g Mb Mb d i2ai vN0 vcai i1ai i2ai v ga vN
Mb Lb + nL g Mb · i2bi = v N0 + vcbi + Rd i1bi − Rd i2bi − vgb − v N (6)
Mb Mb Lb n + L g dt i2ci vN0 vcci i1ci i2ci v gc vN
     
v i i
d  cai  1  1ai   2ai 
 vcbi  = −
 i1bi   i2bi  (7)
dt Cf
vcci i1ci i2ci
  T    T    T  
d a1 i 1a1 d a2 i 1a2 d an i 1an
d 1  l pv −  
v pv = d · i − d · i − · · · − d · i (8)
         
b1 1b1 b2 1b2 bn 1bn 
dt nCo 
           
dc1 i1c1 dc2 i1c2 dcn i1cn

As it was mentioned above, the mutual coupling terms of the three-phase inductors have been
taken into account. Equations (1)–(3) indicate the voltage of the inductor in the phase a, b and c,
respectively, in the inverter side of the LCL filter. Similar equations represent the voltage of the
three-phase inductor in the grid side of the inverter. In addition, the currents of all the inverters are
equal, as expressed by (4), because this will be one of the goals of the control stage.
The PV system has been modelled in a synchronous reference frame (SRF) in which the grid
voltage is aligned with the d-axis. This approach allows independent control of the active power with
the d-channel current, while the reactive power is controlled with the q-channel current. Regulating the
o-channel current, which represents the zero-sequence component, it may be controlled the circulating
current. The equations of the system in the SRF (11)–(14) have been obtained after multiplying
Equations (5)–(8) by the SRF transformation matrix (9) and applying (10) (Park Transformation),
following a similar procedure to the one described in [17]. In these equations ω denotes the grid
angular frequency and x represents any of the average variables. The auxiliary terms Ka , Kb , ILa and
ILb are defined by (15)–(18). It is worth noting that, in [17], a single three-phase three-wire inverter
was studied. In such conditions, the zero-sequence component is null and it was omitted. In addition,
independent inductors (one per phase) were considered, so they are no coupling terms among them:

cos ωt − 2π cos ωt + 2π



r cos ωt 3
3

2
T=  −sin ωt −sin ωt − 2π
3 −sin ωt + 2π
3

(9)
3

√1 √1 √1
2 2 2
       
xd xa xa xd
−1 
 x q  = T · x b  ↔  x b  = T  x q  (10)
      
xo xc xc xo
Energies 2017, 10, 1242 7 of 25

       
i1di i1di ddi vcdi
d dT −1 
dt  i1qi  = − T dt  i1qi  + v pv IL a  dqi  − IL a  vcqi 
     
i1oi i1oi doi vcoi
      (11)
i1di i2di 0
 √
− Rd ILa  i1qi  + Rd ILa  i2qi  − 3v N 0 ILa  0 
    
i1oi i2oi 1
         
i2di i2di vcdi i1di i2di
d dT −1 
dt  i2qi  = − T dt  i2qi  + ILb  vcqi  + Rd ILb  i1qi  − Rd ILb  i2qi 
       
i2oi i2oi vcoi i1oi i2oi
    (12)
v gd 0
 √
− ILb  v gq  − 3(v N 0 − v N ) ILb  0 
  
v go 1
       
vcdi i 2di i 1di i 2di
d dT −1  1 
 vcqi  = −T  i2qi  +  i1qi  −  i2qi  (13)
    
dt dt Cf
vcoi i2oi i1oi i2oi
    T      T  
dd1 i1d1 dd2 i1d2
d 1 
  −1  −1    −1  −1 
dt v pv = nCo  l pv −  T  dq1   · T  i1q1  −  T  dq2   · T  i1q2 
  
do1 i1c1 do2 i1o2
   T   (14)
ddn i1dn 
− . . . −  T −1  dqn  · T −1  i1qn 
    

don i1on

Ka = ( L a + 2Ma )( L a − Ma ) (15)



Kb = Lb + nL g + 2Mb Lb + nL g − Mb (16)
 L +M
−M −M

a a a a
Ka Ka Ka
−M L a + Ma
−M
 a a

IL a = 
 Ka Ka Ka 
 (17)
Ma Ma L a + Ma
− Ka − Ka Ka

Lb +nL g + Mb
 
Kb −M
Kb
b
−M
Kb
b
 
Lb +nL g + Mb
ILb = 

−MKb
b
Kb −M
Kb
b 
 (18)
Lb +nL g + Mb
 
−MKb
b
−MKb
b
Kb

Taking into account (19) and (20) and solving, the equations of the average model in the SRF are
obtained, following (21)–(24):  
− 1 0 −ω 0
dT
T = ω 0 0  (19)
 
dt
0 0 0
 T  T
A −1 B A −1 = B T A −1 A −1 = B T (20)
Energies 2017, 10, 1242 8 of 25

        
i1di 0 −ω 0 i1di ddi vcdi
d
dt  v1qi  = −  ω 0 0  i1qi  + v pv ILa  dqi  − ILa  vcqi 
       
v1oi 0 0 0 i1oi doi vcoi
      (21)
i1di i2di 0
 √
− Rd ILa  i1qi  + Rd ILa  i2qi  − 3v N 0 ILa  0 
    
i1oi i2oi 1
        
i2di 0 −ω 0 i2di vcdi i1di
d
dt  v2qi  = −  ω 0 0  i2qi  + ILb  vcqi  + Rd ILb  v1qi 
       
v2oi 0 0 0 i2oi vcoi v1oi
      (22)
i2di v gd 0
 √
− Rd ILb  v2qi  − ILb  v gq  − 3(v N 0 − v N ) ILb  0 
    
v2oi v go 1
        
vcdi 0 −ω 0 i2di i1di i2di
d 1 
 vcqi  = − ω 0 0  i2qi  +  i1qi  −  i2qi  (23)
      
dt Cf
vcoi 0 0 0 i2oi i1oi i2oi
  T    T    T  
d d1 i 1d1 d d2 i 1d2 d dn i 1dn
d 1 l pv −  
v pv =  dq1  · i1q1  −  dq2  · i1q2  − . . . −  dqn  · i1qn  (24)
         
dt nCo  
do1 i1c1 do2 i1o2 don i1on

2.2. Development of the Small-Signal Model in the State Space

The small-signal equations of the system are obtained by perturbing the average variables around
an operating point. In Equation (25), x represents each one of the variables of the averaged model,
X is the operating point and x̂ is the small-signal value of the variable x. After applying (25) to the
SRF Equations (21)–(24), the small signal equations of the model in the SRF frame have been obtained,
following (26)–(33). Note that the model is quite complex, so that some auxiliary matrices (34)–(43)
have been defined to simplify the expressions of A, B, C and D. Note that the input vector U contains
both control variables (the duty cycles of each inverter leg) and disturbances (the d and the q terms of
the grid voltage):
x = X + x̂ (25)
d
X = A · X + B ·U (26)
dt
Y = C · X + D ·U (27)
" #T
î1d1 î1q1 î1o1 î2d1 î2q1 î2o1 ... î1dn î1qn î1on î2dn î2qn î2on
X=Y= (28)
... ... v̂cd1 v̂cq1 v̂co1 ... v̂cdn v̂cqn v̂con v̂ pv
h iT
U= dˆd1 dˆq1 dˆo1 · · · dˆdn dˆqn dˆon v̂ gd v̂ gq v̂ go (29)
Energies 2017, 10, 1242 9 of 25

··· ···
 
A11 0 0 A21 0 0 A31

 0 A12 ··· 0 0 A22 ··· 0 A32 

 .. .. .. .. .. .. .. .. .. 

 . . . . . . . . . 

0 0 ··· A1n 0 0 ··· A2n A3n
 
 
 
A=
 A41 0 ··· 0 A51 0 ··· 0 0 
 (30)

 0 A42 ··· 0 0 A52 ··· 0 0 

 .. .. .. .. .. .. .. .. .. 

 . . . . . . . . . 

0 0 ··· A4n 0 0 ··· A5n 0
 
 
−K pv
A61 A62 ··· A6n 0 0 ··· 0 nCo
 
B11 0 ··· 0 B21

 0 B12 ··· 0 B22 

 .. .. .. .. .. 

 . . . . .


···
 

 0 0 B1n B2n 

B=
 0 0 ··· 0 0 
 (31)

 0 0 ··· 0 0 

 .. .. .. .. .. 

 . . . . .


···
 
 0 0 0 0 
B31 B32 ··· B3n 0
C=I (32)

D=0 (33)
− Rd ( LKa +a Ma ) Rd Ma Rd Ma Rd ( L a + Ma )
− RdKM − RdKM
 a a

ω+ Ka Ka Ka a a

 −ω + RdKMa a − Rd ( LKa +a Ma ) Rd Ma
− RdKM a Rd ( L a + Ma )
− RdKM a
 
Ka a Ka a

 


Rd Ma
Ka
Rd Ma
Ka − Rd ( LKa +a Ma ) − RdKM
a
a
− RdKM
a
a Rd ( L a + Ma )
Ka


A1i =  Rd ( Lb +nLg + Mb )
 Rd ( Lb +nL g + Mb )
 (34)
 Kb − RdKM
b
b
− RdKM
b
b
− Kb ω+ R d Mb
Kb
R d Mb
Kb


 
R d Mb Rd ( Lb +nL g + Mb ) Rd ( Lb +nL g + Mb )
− Kb − RdKM b
−ω + R d Mb
− R d Mb
 
 Kb b Kb Kb Kb 
Rd ( Lb +nL g + Mb ) Rd ( Lb +nL g + Mb )
− RdKM
b
b
− RdKM
b
b
Kb
R d Mb
Kb
R d Mb
Kb − Kb

− La + Ma Ma Ma
 
Ka Ka Ka
 
Ma

 Ka − La +
Ka
Ma Ma
Ka


 
Ma Ma

 Ka Ka − La +
Ka
Ma 

A2i =  (35)
 
Lb +nL g + Mb
−M −M
b b

 
 Kb Kb Kb 
Lb +nL g + Mb
 

 −M
Kb
b
Kb −M
Kb
b 

 
Lb +nL g + Mb
−M
Kb
b
−M
Kb
b
Kb

d ( L a + Ma )− Dq Ma − Do Ma
 D 
Ka
 
 − Dd Ma + Dq ( L a + Ma )− Do Ma 
Ka
 
 
Dd M a − Dq M a + Do M a
 
A3i =   (36)
 Ka 
0
 
 
 
 0 
0
Energies 2017, 10, 1242 10 of 25

1
− C1f
 
Cf 0 0 0 0
1
− C1f
 
A4i =  0 Cf 0 0 0  (37)
 
1
0 0 Cf 0 0 − C1f
 
0 ω 0
A5i =  −ω 0 0  (38)
 
0 0 0
h i
Dd D − Do
A6i = − nCo
− nCqo − DonCo
0 0 0 (39)
 
Vpv ( L a + Ma ) Vpv Ma Vpv Ma
Ka − Ka − Ka
 
Vpv Ma Vpv ( L a + Ma ) Vpv Ma
− Ka − Ka
 

 Ka 

V M Vpv Ma Vpv ( L a + Ma )
− pvKa a − Ka
 
B1i = 
 Ka

 (40)
0 0 0
 
 
0 0 0
 
 
0 0 0
 
0 0 0

 0 0 0 


 0 0 0 

B2i =  − Lb +nL g + Mb Mb Mb (41)
 

 Kb Kb Kb 
 Mb Lb +nL g + Mb Mb 

 Kb − Kb Kb


Mb Mb L +nL + M
Kb Kb − b Kbg b
h i
Id Iq Io
B31,2, ...,n−1 = − nC − nCo − nCo
(42)
o
h i
Id Iq −(n−1) Io
B3n = − nC − nCo − nCo
(43)
o

The photovoltaic field has been modelled by linearization of the current curve of the panels
around an operation point close to the maximum power point (MPP). In Equation (44) the linearization
of the current supplied by a PV field composed by np parallel branches (or strings) and ns series panels
per string is shown, following the procedure that was described in [17]. Ipv and Vpv are the current and
the voltage, respectively, of the PV field at the considered MPP:

n p I pv 
k pv = − ·  (44)
ns Vpv  MPP

The expressions of the variables at the operating point are represented in Table 1. To simplify the
calculation of these terms, the mutual coupling terms of the three-phase inductors have been neglected
and the capacitors of the filters have been not considered, so that the currents in the grid side of the
filters (I1dqoi ) and in the converter side (I2dqoi ) agree (Idqoi ≈ I1dqoi ≈ I2dqoi ). In addition, the expressions
of the operating point are obtained by taking into account the control objectives, so that currents in the
active channel of each inverter are considered of the same value because an equally distribution of
the power generation between the inverters is desired. This condition will be imposed by the control
stage. To obtain null reactive power and avoid circulating currents, the reactive and the zero-sequence
components of the currents must be null.
Energies 2017, 10, 1242 11 of 25

Table 1. Simplified expression of the operating point values.

Variable Variable Variable

Vgd VgRMS (phase-phase) Vgq 0 Vgo 0
Vpv I pv
Id n·V gd Iq 0 Io 0
Vgd ωId ( L a +nL g + Lb )
Dd Vpv Dq Vpv
Do 0

3. Description of the Proposed Control Architecture for n Parallel PV Inverters

3.1. Overview of the Control Stage

Figure 7 shows the proposed control structure for the previosuly described PV system. It is
composed of n d-channel and n q-channel current control loops (in both cases, one per inverter) and
(n − 1) o-channel current controllers. It is worth noting that more than one circulating current path
exists in the schematic of two parallel inverters. Instead of looking at all individual circulating currents,
the zero-sequence current can be used to represent the overall circulating current. The zero-sequence
current is defined as the sum of all phases’ currents in one inverter. For instance, taken the inverter 1:
io1 = ia1 + ib1 + ic1 . It is obvious that io1 = 0 if only one inverter is connected, because it generates a set
of balanced currents (ia1 + ib1 + ic1 = 0). Taking n = 2 (two modules connected in parallel), io1 = −io2 ,
so that both currents can be controlled by acting only on one of them. In general, for n inverters in
parallel, n − 1 controllers are needed to regulate all the o-channel currents.
The PV field voltage is controlled by a single outer voltage loop, which calculates the reference
signal for the d-channel inner controllers. The reference for the voltage loop would be calculated by a
Maximum Power Point Tracking (MPPT) algorithm, whose study is beyond the scope of this work.
With this approach, the d-component of the current allows the control of the active power of the system
and the q-channel current control loop allows the control of the reactive power. Unity power factor is
usually desired to take profit of the full power capacity of inverters to generate active power, so the
reference of the q-channel control loop is normally zero. Finally, each power module has its own Space
Vector Modulator (SVM), which works from the duty cycles calculated by the current controllers.
It is worth to point out that the novelties of the proposed control stage with regard to the
conventional one are, basically, two. On one hand, the zero sequence component of three-phase
inverters are normally supposed to be zero, so that this component is not controlled. In the proposed
solution, the zero sequence components are controlled with a reference signal io * = 0 to avoid the
appearance of circulating currents among inverters. On the other hand, 2D-SVM are normally used
for three-phase three-wire systems. However, 2D-SVM can only operate in the αβ plane so it can not
generate zero sequence components that are outside that plane, as it is the case of the zero-sequence
componentes. To overcome this limitation, the use of a 3D-SVM is proposed and detailed in the
next section. Note that only (n − 1) 3D modulators are needed, the same number as o-channel
current controllers.
 Energies 2017, 10, 1242 12 of 25

section. Note that only (n − 1) 3D modulators are needed, the same number as o-channel current
Energies 2017, 10, 1242 12 of 25
controllers.

Schematicofof
Figure7.7.Schematic
Figure the
the proposed
proposed Control
Control Stage.
Stage.

Inany
In any system
system represented
representedin inthe
thesynchronous
synchronousreference
reference frame, there
frame, areare
there coupling terms
coupling between
terms between
the d- and q-axis. The small-signal model presented in Section 2 show the presence of
the d- and q-axis. The small-signal model presented in Section 2 show the presence of such coupling such coupling
terms. Equations (47)–(48) express the decoupling terms between both axis, which have been
terms. Equations (47)–(48) express the decoupling terms between both axis, which have been caculated
caculated by necglecting the filter capacitors, so that the currents in the grid side of the filters (I1dqoi)
by necglecting the filter capacitors, so that the currents in the grid side of the filters (I1dqoi ) and in the
and in the converter side (I2dqoi) agree (Idqoi ≈ I1dqoi ≈ I2dqoi). Assuming, in adition, that ( ) ( ) = 0,
converter side (I2dqoi ) agree (Idqoi ≈ I1dqoi ≈ I2dqoi ). Assuming, in adition, that v̂ pv (s)v̂ gd (s) = 0, it is
it is obtained (45). Being εidi the current error in the d-channel, and PIi a proportional-integral
obtained (45). Being εidi the current error in the d-channel, and PIi a proportional-integral regulator, the
regulator, the control action on the d-channel is expressed as in Equation (46). From equations (45)–
control action on the d-channel is expressed as in Equation (46). From Equations (45)–(46) it is obtained
(46) it is obtained the decoupling term Kqd. The Kdq term is calculated following the same procedure.
the decoupling term Kqd .are
These decoupling terms
The Kdq from
valid
term dc
is calculated following the same procedure. These decoupling
up to medium frequencies:
terms are valid from dc up to medium frequencies:
( )· + + + · ̂ ( )= + + s· ̂ ( )
(45)
ˆ



ddi (s)·Vpv + ω L a + nL g + Lb ·îqi (s) = L a + nL g + Lb s·îdi (s) (45)
( )= (s) ̂ ( ) + · ̂ ( )
dˆdi (s) = PIi (s)ε̂ idi (s) + Kqd ·îqi (s) (46) (46)

ω (L a ++ nL g ++ Lb)


=
k qd = − − (47) (47)
Vpv
ω (L a ++ nL g ++ Lb)


k dq == (48) (48)
Vpv
Equations
Equations(49)–(51)
(49)–(51)represent
representthethecurrent control
current controllooploop in theind- the
gainsgains q- and
d- o-channels, respectively.
q- and o-channels,
respectively.
In In theRequations,
the equations, s is the R
current
s is the current
sensor sensor
gain, PI i gain,
(s) arePI (s) are proportional-integral
proportional-integral
i regulators
regulators and D(s)
and D(s) represents a delay equivalent to one switching period, which has
represents a delay equivalent to one switching period, which has been calculated by means of been calculated by means
aof a second-order
second-order Padé
Padé approximant.
approximant. The
The dutycycle
duty cycletotocurrent
currenttransfer
transferfunctions
functions areare obtained
obtained from
from the
the state
state spacespace model
model of the
of the system
system by means
by means of MATLAB (R2015b).
of MATLAB TMTM (R2015b).Note Notethat
that the
the current
current sensors
sensors are
are placed
placed oninverter
on the the inverter
side ofside
theofLCLthe filter.
LCL filter. Locating
Locating the sensors
the sensors on this onside
thisofside
the of the inverter
inverter improves
improves the robustness of the system instead of placing the sensors
the robustness of the system instead of placing the sensors on the grid side [18]: on the grid side [18]:
̂ ( )
( )= · ( ) · ( ) ·î (s) (49)
Tidi (s) = Rs · PIi (s)· D (s)· i1di ( ) (49)
ˆ
ddi (s)

îi1qi (s)
Tiqi (s) = Rs · PIi (s)· D (s)· (50)
dˆqi (s)
 ̂ ( )
( )= · ( )· ( )· (50)
( )

Energies 2017, 10, 1242 ̂ ( ) 13 of 25

( )= · ( )· ( )· (51)
( )
The voltage control loop gain follows Equation (52). In this î (equation,
s) β is the voltage sensor gain
Tioi (s) = Rs · PIi (s)· D (s)· i1oi (51)
and PIv(s) is a proportional-integral regulator: dˆoi (s)
̂ ( ) (52).
The voltage control loop gain follows Equation ( )In this equation, β is
( the
) voltage sensor gain
( )= · ( )· · = · ( )· (52)
̂
and PIv (s) is a proportional-integral regulator: ( ) ̂ ( ) ̂ ( )

î1di (s) v̂o (s) v̂o (s)

β· PIv (s)·
(s) =Modulator
3.2. Three-Dimension SpaceTviVector · = β· PIv (s)· (52)
îdre f (s) î1di (s) îdre f (s)
A three-phase converter has eight possible commutation vectors. In Table 2 the phase to neutral
3.2. Three-Dimension
voltages Space
of the different Vector Modulator
switching vectors are expressed taking into account that the low frequency
components of the voltage between the middle point of the dc-link and the neutral of the mains are
A three-phase converter has eight possible commutation vectors. In Table 2 the phase to neutral
zero, so it can be considered that the neutral point of the grid and the middle point of the dc-link are
voltages of the different switching vectors are expressed taking into account that the low frequency
virtually connected at the frequency of the fundamental line voltage [19].
components of the voltage between the middle point of the dc-link and the neutral of the mains are
zero, so it can be considered that the neutral point of the grid and the middle point of the dc-link are
Table 2. Commutation vectors of 3D-SVM.
virtually connected at the frequency of the fundamental line voltage [19].
vφ v0-000 v1-100 v2-110 v3-010 v4-011 v5-001 v6-101 v7-111
1 1 1 2. Commutation
Table 1 vectors1of 3D-SVM. 1 1 1
− − − −
2 2 2 2 2 2 2 2
1 1 1 1 1 1 1 1
vϕ − 2 v0-000 − v1-100
2 2 v2-110 2v3-010 2v4-011
−
v5-001
2 v−6-101
2 v7-111
2
1 1 1 1 1 1 11 1 1
v A − − 21 Vpv − 12 Vpv − 21 Vpv −− 12 Vpv − 12 Vpv − 12 Vpv V
22 pv V
2 2pv
2 1 2 1 2 2 21 2
− Vpv 1 12 Vpv 112 Vpv − 112 Vpv −112 Vpv 1
2 − 2 Vpv
vB 2 Vpv 2 Vpv0
0 12 − 1 2
− 1 3 −1
vC − 2 Vpv 3 1V
− 1
√6− 2 Vpv −
√62 Vpv
1 1
√ 2 pv 2 Vpv
√ 2√6Vpv √6
2 Vpv 2 Vpv
1 1 1 1 1 1 √1 1Vpv
vα 0 0 0 2/3 Vpv √ Vpv − √ Vpv − 02/3 Vpv −
− √ Vpv − 0 0
6 6 6 6
√2 1 √21 √21 √2
v β −√3 0 −1 0 √
1 2 Vpv √
−1 2 Vpv 1 0 −−1 Vpv
√ − √11 Vpv 0
√3
√ 2 2 √
vo 2 − 2 3 Vpv 2√3−
√1 Vpv 1
2√3√ Vpv −
√1 Vpv
2√3 1
2√3
√ Vpv −√1 Vpv
2√3 1
2√3
√ Vpv 32
2 3 2 3 2 3 2 3 2 3 2 3 2 Vpv

Figure 8 represents the switching vectors in αβ0 (Clarke) coordinates. The switching vectors
Figure
projected 8 represents
onto thedefine
the α, β plane switching vectorsEach
six prisms. in αβ0 (Clarke)
prism coordinates.
contains The switching
two tetrahedrons vectors
defined by two
projected onto the α, β plane define six prisms. Each prism contains two tetrahedrons
non-zero switching consecutive vectors and the neutral vectors v0 and v7 alternatively. defined by two
non-zero switching consecutive vectors and the neutral vectors v0 and v7 alternatively.

III II I
IV V VI v7
v7
v2 v2
v4 β
v6 v4
α v6
v3
v1
v5 v3

v0 v1
v5
v0
(a) (b)
Figure 8. (a)
Figure 8. (a)Spatial
Spatial distribution
distribution ofcommutation
of the the commutation
vectors;vectors; (b)distribution
(b) Spatial Spatial distribution of the
of the tetrahedrons.
tetrahedrons.

The tetrahedron that contains the reference vector must be calculated to obtain the switching
vectors that must be applied [20]. It is necessary to calculate the time of application of the vectors,
projecting the reference vector to the switching vectors that delimit the tetrahedrons.
Energies 2017, 10, 1242 14 of 25

The tetrahedron that contains the reference vector must be calculated to obtain the switching
vectors that must be applied [20]. It is necessary to calculate the time of application of the vectors,
Energies 2017,
projecting the10,reference
1242 vector to the switching vectors that delimit the tetrahedrons. 14 of 25
Equations (53)–(56) define the time of application of each switching vector. The residual time in
every switching period is applied equally to the neutral vector v0 and v7:
Equations (53)–(56) define the time of application of each switching vector. The residual time in
every switching period is applied equally to the neutral vector v0 and v7 :
1
 = (53)
dn T
1 h i
 d n +1  = v n v n +1 v 0 (53)
 
Vpv
d0 = · (54)

tvn = Ts ·dn (54)

= ·
(55)
tvn+1 = Ts ·dn+1 (55)

tv= ·
0 = Ts · d v0 (56)
(56)
Thedistribution
The distributiontime
timeofofthe
thecommutation
commutationvectors
vectorsfollows
followsnormally
normallya asymmetric
symmetricdistribution.
distribution.
Then, the time of application of each non-neutral vector is distributed in two periods
Then, the time of application of each non-neutral vector is distributed in two periods before and before andafter
after
the v vector.
the v0 vector.
0 It is important to reduce the number of commutations in order to decrease
It is important to reduce the number of commutations in order to decrease power lossespower losses
in semiconductors.
in semiconductors. Table Table3 3represents
representsthe
thevectors
vectorssequence
sequencethat
thatmust
mustbebeapplied
appliedinineach
eachprism.
prism.

Commutationvectors
Table3.3.Commutation
Table vectorssequence
sequenceinineach
eachprism.
prism.

PrismPrism Commutation
Commutation Vectors
VectorsSequenceSequence
I v7-v2-v1-v0-v1-v2-v7
I v7 -v2 -v1 -v0 -v1 -v2 -v7
II II vv7-v-v2-v-v 3-v0-v3-v2-v0
7 2 3 -v0 -v3 -v2 -v0
III III vv7-v 4 -v 3-v0-v3-v4-v7
7 -v4 -v3 -v0 -v3 -v4 -v7
IV IV vv7-v
7 4 5-v
-v4-v-v 0-v5-v
5 -v 0 -v54-v-v47-v7
V V vv7-v
7 6 5 0 5 -v67-v7
-v6 -v -v
5 -v -v
0 -v -v
5 -v 6-v
VI VI vv7-v
7 -v6-v6 -v
1-v1 -v
0-v01-v-v16-v
-v67-v7

4.4.Application
Applicationtotoa a2 2MW
MWPV
PVSystem
SystemComposed
Composedby
byFour
Four500
500kW
kWModules
ModulesininParallel
Parallel
The
Theproposed
proposed control structurehas
control structure hasbeen
been applied
applied to ato2 MW
a 2 MW PV system
PV system composed
composed by fourby four
modules,
modules,
each oneeach one
of 500 of as
kW, 500itkW, as it isinshown
is shown Figurein9. Figure
In this 9. In thisitsection,
section, it is presented
is presented the MATLAB
the MATLAB TM script that
TM

script that developed

has been has been developed
to study theto study
dynamic theresponse
dynamicofresponse of the
the system. system.
Then, someThen, some
results results areto
are presented
presented to validate
validate the concept.the concept.

PCC

+
Vpv
-

Figure 9. 9.
Figure Line diagram
Line ofof
diagram the 2 MW
the centralized
2 MW PVPV
centralized inverter under
inverter study.
under study.

Table4 4indicates
Table indicatesthe
thevalue
valueofofthe
theparameters
parameterscorresponding
correspondingtotoone
oneofofthe
theinverters
invertersunder
understudy.
study.
Energies 2017, 10, 1242 15 of 25

Table 4. Parameters of the Parallel Inverters.

Parameter Nominal Value Parameter Nominal Value

Vg -RMS (line voltage) 230 V Ma −20 µH
Vpv (MPPT range) [650–820] V Mb −10 µH
Vdc max 1000 V Cf 500 µF
Pn 500 kW Rd 0.1 Ω
Co 15 mF fsw 2 kHz
La 80 µH Rs 1 V/A
Lb 40 µH β 1 V/V

An issue that is especially relevant for the analysis and design of high power PV inverters is
the value of the grid impedances, Lg , which strongly depends on the Point of Common Coupling
(PCC) where the PV installation is connected to the grid. To analyze this question, let us define the
short-circuit factor Rsc as the relation between the short-circuit power Ssc at the connection point and
the nominal power Sn of the PV system, following Equation (57):

Ssc
Rsc = (57)
Sn

Assuming that the short-circuit impedance is mainly inductive, the grid inductance (i.e., the
short-circuit inductance) Lg can be expressed by Equation (58):

3·Vg2
Lg ≈ (58)
ω · Rsc Sn

According to Rsc , the nature of the grid can be classified as weak (for Rsc low) and strong (for
Rsc high). In the case under study, three values of Lg have been considered to analyze the dynamic
response of the PV system with the proposed control stage, as it has been summarized in Table 5.

Table 5. Summary of the considered values for the grid inductance Lg .

Nature of the Grid Rsc Lg

Weak 5 50 µH
Normal 20 12.7 µH
Strong 100 2.5 µH

4.1. Dynamic Analysis of the Control Loops

In this section it is studied the stability of the control loops of the whole system. The MATLABTM
script employed to obtain the dynamics of the control loops is presented in Figures 10 and 11. Figure 10
shows the code that allows the study of the state-space model, taking into account the decoupling terms.
Figure 11 shows the part of the code that it has been used to calculate the current and voltage control
loop gains. Figure 12 illustrates the interconnections that are described in the MATLABTM script.
In Figure 10, firstly, the section “parameters definition” describes the parameters of the system
under study, as they were expressed in Table 4. Then, the operating point variables and the developed
state-space equations are expressed, following Table 1 and (26)–(44), respectively. Note that the
matrices A and B are not detailed because their size is excessive to this paper. They must be added
to the script following Equations (30)–(31). Then, the “ss” function of MATLABTM is employed to
generate the state-space model of the system, starting from the A, B, C and D matrices. The obtained
state-space model, “G_without_dec”, is not decoupled. To add the decoupling terms between the
d- and q-channels that were presented in Equations (47)–(48), a decoupling matrix, “Dec”, has been
generated. After combining the decoupling matrix and the model of the system by means of a positive
“feedback” function, the decoupled state-space model, “G_dec”, is obtained. Figure 11 shows the
Energies 2017, 10, 1242 16 of 25

process to calculate the current and voltage control loop gains. In this part of the code, it is defined
Energies 2017, 10, 1242 16 of 25
a delay equivalent to one switching period, which has been calculated by means of a second-order
Padé approximant.
approximant. After
After that, it isthat, it is presented
presented the expression the expression of the proportional-integral
of the proportional-integral current
current regulators
regulators (59) and Equations (49)–(51) are added to the script. Note
(59) and Equations (49)–(51) are added to the script. Note that the duty cycle to current transfer that the duty cycle to current
transfer functions are obtained with the “tf” function of MATLAB.
functions are obtained with the “tf” function of MATLAB. This function generates the transfer This function generates the transfer
function of
function of the
the selected input,dˆdi , dˆ,qi , dˆoi, , to, output
selectedinput, to outputvariables,
variables, ̂ ,,î1oi
î1di , î1qi ̂ ,,respectively,
̂ , respectively,
starting from
starting
the decoupled state-space model of the system “G_dec”. Finally,
from the decoupled state-space model of the system “G_dec”. Finally, the “margin” command the “margin” command deploys the
Bode plotthe
deploys of the
Boderespective
plot of transfer function,
the respective as Figure
transfer 13 shows,
function, as and
Figure calculates
13 shows,the crossover frequency
and calculates the
and the gain and the phase margins. Because of the components in
crossover frequency and the gain and the phase margins. Because of the components in each inverter each inverter are considered to
have the same value, the dynamic response is the same in the d-, q-
are considered to have the same value, the dynamic response is the same in the d-, q- and o-channel and o-channel of all inverters.
Forallthis
of reason, For
inverters. onlythis
onereason,
inverter needs
only oneto be studied.
inverter needs to be studied.

Figure 10. MATLAB

Figure 10. MATLABTM script.
TM
script.Definition
Definitionof
ofthe
thestate-space
state-spacemodel.
model.
Energies 2017, 10, 1242 17 of 25
Energies 2017, 10, 1242 17 of 25
Energies 2017, 10, 1242 17 of 25

Figure
11.11. MATLAB script.
script. Current
Current and
andvoltage control loop gains.
TM
Figure
Figure MATLAB
11. TMTMscript.
MATLAB Current and voltage
voltage control
control loop
loop gains.
gains.

G_CL_I
iidref1 G_CL_I
dref1 G_PI
iiqref1 G_PI
qref1
iioref1 G_dec
G_dec
oref1
...
...
iidrefn i1d1 Y
+ PI_MATRIX + i1d1 Y
drefn
iiqrefn +- PI_MATRIX ++ i1q1
- + i1q1
qrefn
iiorefn i1o1
i1o1
orefn ...
vvgd G_without_dec
i1dn
...
gd G_without_dec
vvgq i1dn
i1qn
gq i1qn
vvgo i1on
go i1on
...
...

in_Dec out_Dec
in_Dec Dec out_Dec
Dec

in_CL_I out_CL_I
in_CL_I K out_CL_I
K

Figure 12. MATLABTM scheme of interconnection.

Figure MATLABTM
Figure 12.12.
MATLAB TM scheme of interconnection.
scheme of interconnection.

Before calculating the voltage control loop gain, the closed current control loops are defined. First,
a “PI_MATRIX” containing the expressions of the regulators for the three current loops of each inverter,
Energies 2017, 10, 1242 18 of 25

is generated and it is connected to the state-space model. Note that this matrix has n d-, n q-, and n − 1
o-current regulators expressed as “PI_M” and “PI_Mn”. The “PI_MATRIX” is connected in series to
the state-space model. There is a value of “1” for the “PI_Mn” and the “PI_MVg” values where a
regulator must not be connected to the disturbance inputs, dˆon , v̂ gd , v̂ gq and v̂ go . A sensor current gain
matrix, “K”, is also created to take into account the gain of the sensors. Then, a negative “feedback”
function is used to close the current loops, by linking the duty cycle (inputs) with the sensed currents
(outputs), of “G_PI”. As a result, it is obtained the model of the system after closing the current loops,
“G_CL_I”.
Finally, it is obtained the sensed current to voltage transfer function, “GVpv_idref1”. Then, the
expression of the proportional-integral voltage regulator (Equation (60)) and the one of the voltage
current loop gain (Equation (52)) are added. The “margin” command is applied to evaluate the stability
of the voltage loop.
The vectors “in_Dec” and “in_CL_I” contains the inputs of “G_without_dec” and “G_PI”,
respectively, that are involved in the feedback loop. Similarly, “out_Dec” and “out_CL_I” specifies the
outputs that are used for feedback. The resulting matrix after applying the “feedback” command, has
the same inputs and outputs as the initial system.
It is worth noting that the expressions of the regulators have been chosen by means of an iterative
trial and error procedure, using as design criteria the following:

• The crossover frequency of the current loops gain should be much higher than the fundamental
frequency of the grid. In addition, it should be much lower than the switching frequency to avoid
large-signal instabilities.
• The crossover frequency of the voltage loop should be much lower than the one of the current
loops (typically around a factor 1/10).
• In both cases, proper stability margins (PM > 45◦ and GM > 6 dB) should be obtained.

Equations (59) and (60) express the chosen proportional-integral regulators for the current control
loops and for the voltage control loop, respectively. As it will be shown later, the dynamics of the three
current loops (d, q and o) are similar, so their corresponding controllers can have the same expression:

0.1
PIi (s) = 0.00025 + (59)
s
100
PIv (s) = −20 − (60)
s
Figure 13 represents the Bode plots (dB, deg) of the three current loop gains, Tid (jw) (49), Tiq (jw) (50)
and Tio (jw) (51) for one of the power modules. Figure 13 also shows the Bode plots of the voltage loop
gain, Tv (jw) (52). In all cases, it has been considered that the PV voltage can vary inside the MPPT
range (from 650 V to 820 V). In addition, three representative values of the grid inductance have been
considered, following Table 5, to take the nature of the grid into account. As can be seen in Figure 13,
the dynamics of the system are sensitive to the considered variations of Vpv and Lg , but the stability is
guaranteed in all cases. Concretely, the crossover frequency of the current loops ranges from 90 Hz to
300 Hz, having a phase margin PM = 50 and a gain margin GM = 5 db in the worst case. Regarding the
voltage loop gain, the crossover frequency ranges from 22 Hz to 27 Hz and the phase margin is higher
than 90 degrees.
Energies 2017, 10, 1242 19 of 25
Energies 2017, 10, 1242 19 of 25

Weak grid Weak grid

Normal grid Normal grid
Strong grid Strong grid
Vpv=820 Vpv=820
Vpv=750 Vpv=750
Vpv=650 Vpv=650

(a) (b)

Weak grid
Weak grid Normal grid
Normal grid Strong grid
Strong grid Vpv=820
Vpv=820 Vpv=750
Vpv=750 Vpv=650
Vpv=650

(c)
(d)

Figure 13. Control loops: (a) Current control loop gain in d-channel; (b) Current control loop gain in
Figure 13. Control loops: (a) Current control loop gain in d-channel; (b) Current control loop gain in
q-channel; (c) Current control loop gain in o-channel; (d) Voltage control loop gain.
q-channel; (c) Current control loop gain in o-channel; (d) Voltage control loop gain.
4.2. Results
4.2. Results
To validate the small-signal design in the large signal sense, the performance of the PV system
hasvalidate
To been evaluated by means of design
the small-signal PSIMTM in (10.0)
thesoftware. Figure
large signal 14b shows
sense, the evolution of the
the performance thecurrent
PV system
(phase
has been a) of the four
evaluated by power
meansmodules
of PSIM TM (10.0)
connected insoftware.
parallel, as Figure
the generated power the
14b shows ranges from zeroof the
evolution
to 1(phase
current MW, Figure
a) of 14a,
the and
fourworking
power with
modulesa conventional
connected control scheme as
in parallel, that
thedoesn’t controlpower
generated the zero-
ranges
sequence components of the currents. It is worth noting that it is desired an equal power distribution
from zero to 1 MW, Figure 14a, and working with a conventional control scheme that doesn’t control
among the modules as they connect to the system. In the Introduction (Figure 3), it was explained
the zero-sequence components of the currents. It is worth noting that it is desired an equal power
that this sequence of connection allows having a good efficiency in all the range of power, reducing
distribution among the modules as they connect to the system. In the Introduction (Figure 3), it
considerably the power loss when the generated power is low. Figure 14c depicts the zero-sequence
was explained that
current of the this sequence
inverters. of activation
Before the connection allows
of the having
second a good
inverter, whenefficiency
the first onein isallmanaging
the range of
power, reducing considerably the power loss when the generated power is low. Figure
all the generated power, there is no circulating current. However, after the activation of the second 14c depicts the
zero-sequence current ofcurrent
inverter a circulating the inverters.
appears Before
betweenthe activation
both inverters.ofThe
thesame
second inverter,
occurs when
after the the first
activation of one
the inverters
is managing 3 and
all the 4.
generated power, there is no circulating current. However, after the activation of
the second inverter a circulating current appears between both inverters. The same occurs after the
activation of the inverters 3 and 4.
Energies 2017, 10, 1242 20 of 25
Energies 2017, 10, 1242 20 of 25

1000K Inv. 2 activation

Inv. 3 activation
800K Inv. 4 activation
Power (W)

600K

400K

200K

0K

0 0.2 0.4 0.6 0.8 1

Time (s)
(a)
1K Inv. 2 activation
i1a1
Inv. 3 activation i1a2
0.5K Inv. 4 activation i1a3
i1a4
Current (A)

0K

-0.5K

-1K

0 0.2 0.4 0.6 0.8 1

Time (s)
(b)
300
Inv. 2 activation io1
200 io2
Inv. 3 activation
Inv. 4 activation io3
100 io4
Current (A)

-100

-200

-300
0 0.2 0.4 0.6 0.8 1
Time (s)

(c)

Figure 14. Activation sequence of the inverters without zero-sequence control: (a) Generated PV
Figure 14. Activation sequence of the inverters without zero-sequence control: (a) Generated PV power;
power; (b) Current grid in the phase A of each inverter; (c) Zero-sequence current of each inverter.
(b) Current grid in the phase A of each inverter; (c) Zero-sequence current of each inverter.
Figure 15 shows the same waveforms, following the same activation sequence of inverters that
Figure 1514,
in Figure but working
shows withwaveforms,
the same the proposedfollowing
control scheme. It canactivation
the same be seen in Figure 15c that
sequence the zero- that
of inverters
sequence currents are null, so there are not circulating currents among inverters.
in Figure 14, but working with the proposed control scheme. It can be seen in Figure 15c It is worth noting
that the
that the power distribution among the modules is much better with the proposed control technique.
zero-sequence currents are null, so there are not circulating currents among inverters. It is worth noting
that the power distribution among the modules is much better with the proposed control technique.
Energies 2017, 10, 1242 21 of 25
Energies 2017, 10, 1242 21 of 25

1000K Inv. 2 activation

Inv. 3 activation
800K Inv. 4 activation
Power (W)

600K

400K

200K

0K

0 0.2 0.4 0.6 0.8 1

Time (s)

(a)

1K Inv. 2 activation
i1a1
Inv. 3 activation i1a2
0.5K Inv. 4 activation i1a3
Current (A)

i1a4
0K

-0.5K

-1K

0 0.2 0.4 0.6 0.8 1

Time (s)

(b)
300
Inv. 2 activation
200 Inv. 3 activation io1
Inv. 4 activation io2
100 io3
io4
Current (A)

-100

-200

-300
0 0.2 0.4 0.6 0.8 1
Time (s)

(c)

Figure 15. Activation sequence of the inverters with zero-sequence control: (a) Generated PV power;
Figure 15. Activation sequence of the inverters with zero-sequence control: (a) Generated PV power;
(b) Grid current in the phase A of each inverter; (c) Zero-sequence current of each inverter.
(b) Grid current in the phase A of each inverter; (c) Zero-sequence current of each inverter.
Figure 16a shows the inverters’ currents in steady state and full load conditions. Before the
activation
Figure 16a of shows
the zero-sequence controllers,
the inverters’ currents there
in is a significant
steady imbalance
state and between
full load the currents
conditions. of the
Before
the inverters and after the activation at t = 2.4 s the currents become equal. In this figure
activation of the zero-sequence controllers, there is a significant imbalance between the currents of it has not
been implemented
the inverters and afterthe thecurrent limitation
activation at t =of2.4
thesinverters, so that
the currents the generated
become equal. power
In thisisfigure
maintained
it has not
at 2 MW, but some of the modules are overloaded while others are underloaded. Figure 16c depicts
been implemented the current limitation of the inverters, so that the generated power is maintained at
the zero-sequence current in each inverter also before and after the activation of the zero-sequence
2 MW, but some of the modules are overloaded while others are underloaded. Figure 16c depicts the
control at t = 2.4 s. The circulating current reaches approximately 30% of the fundamental current in
zero-sequence current
the worst case in the
before each inverterofalso
activation the before
control and afterbecomes
and then the activation
zero. of the zero-sequence control
at t = 2.4 s.
In The circulating
practice, current
the inverters reaches
will deliverapproximately
as maximum their 30%nominal
of the fundamental current
current, so the actualin the worst
power
case before thebe
that could activation
injected toofthe
thegrid
control
would andbe then becomes
lower. zero.
As a result, the power capacity of the PV system
In practice, the inverters will deliver as maximum their nominal current, so the actual power that
could be injected to the grid would be lower. As a result, the power capacity of the PV system would
be degraded. This fact is illustrated by Figure 17. In this figure, the performance of the PV system
Energies 2017, 10, 1242 22 of 25
Energies 2017, 10, 1242 22 of 25

would be degraded. This fact is illustrated by Figure 17. In this figure, the performance of the PV
has been evaluated in similar conditions that in Figure 16, but limiting the inverters’ current to their
system has been evaluated in similar conditions that in Figure 16, but limiting the inverters’ current
nominal value. Note value.
to their nominal that, before the activation
Note that, before the of the proposed
activation control technique,
of the proposed the PV system
control technique, the PV can’t
reachsystem
the nominal power.
can’t reach the Once the power.
nominal zero sequence
Once thecontrollers are activated,
zero sequence controllers the
are circulating current
activated, the
disappears andcurrent
circulating the delivered power
disappears is equally
and the distributed
delivered amongdistributed
power is equally inverters,among
so theinverters,
nominal so2 MW
the can
nominal 2 MW can be supplied.
be supplied.

2.1M
Zero-sequence control activation

2M
Power (W)

1.9M

1.8M

1.7M
2.25 2.3 2.35 2.4 2.45 2.5 2.55
Time (s)

(a)

Zero-sequence control activation

1K
i1a1
i1a2
0.5K
Current (A)

i1a3
i1a4
0K

-0.5K

-1K

2.25 2.3 2.35 2.4 2.45 2.5 2.55

Time (s)

(b)

400
Zero-sequence control activation io1
io2
200
io3
Current (A)

io4
0

-200

-400

2.25 2.3 2.35 2.4 2.45 2.5 2.55

Time (s)
(c)

Figure 16. Activation of the zero-sequence control without current limitation: (a) Generated PV
Figure 16. Activation of the zero-sequence control without current limitation: (a) Generated PV power;
power; (b) Grid current in the phase A of each inverter; (c) Zero-sequence current or circulating
(b) Grid current in the phase A of each inverter; (c) Zero-sequence current or circulating current of
current of each inverter.
each inverter.
Energies 2017, 10, 1242 23 of 25
Energies 2017, 10, 1242 23 of 25

2.1M
Zero-sequence control activation

2M
Power (W)

1.9M

1.8M

1.7M
1.35 1.4 1.45 1.5 1.55 1.6 1.65
Time (s)

(a)
1K Zero-sequence control activation
i1a1
i1a2
0.5K i1a3
i1a4
Current (A)

0K

-0.5K

-1K
1.35 1.4 1.45 1.5 1.55 1.6 1.65
Time (s)

(b)
300
Zero-sequence control activation io1
200 io2
io3
100 io4
Current (A)

-100

-200

-300
1.35 1.4 1.45 1.5 1.55 1.6 1.65
Time (s)
(c)

Figure 17. Activation of the zero-sequence control with current limitation: (a) Generated PV power
Figure 17. Activation of the zero-sequence control with current limitation: (a) Generated PV power
(b) Grid current in the phase A of each inverter; (c) Zero-sequence current or circulating current of
(b) Grid current in the phase A of each inverter; (c) Zero-sequence current or circulating current of
each inverter.
each inverter.
5. Conclusions
5. Conclusions
This work has presented an accurate model of centralized PV transformerless inverters for high
power photovoltaic
This work fields composed
has presented an accurate by model
n parallel modules. Such
of centralized PV atransformerless
power architecture has many
inverters for high
benefits such as modularity, good efficiency from low to high power generation
power photovoltaic fields composed by n parallel modules. Such a power architecture and reduced weight,
has many
volume and costs. However, it has been outlined that significant circulating currents appear among
benefits such as modularity, good efficiency from low to high power generation and reduced weight,
the modules and this degrades the performance of the PV system.
volume and costs. However, it has been outlined that significant circulating currents appear among
To overcome this problem, a control technique that regulates the zero-sequence component of
the modules and this degrades the performance of the PV system.
the inverters’ currents has been proposed. Moreover, it has been pointed that 3D Space Vector
To overcome
Modulators arethis problem,
needed a control
to control technique that
the zero-sequence regulatesofthe
components thezero-sequence
currents. component of the
inverters’ currents has been proposed. Moreover, it has been pointed that 3D Space Vector Modulators
are needed to control the zero-sequence components of the currents.
Energies 2017, 10, 1242 24 of 25

The proposed control stage has been applied to a 2 MW photovoltaic system composed by four
parallel inverters, 500 kW each, that are connected in parallel. The presented results show that,
as expected, the circulating currents become zero after activating the control of the zero-sequence
component of the inverters’ currents. The concept can be easily implemented in PV systems because it
does not add significant extra complexity.

Acknowledgments: This work is supported by the Spanish Ministry of Economy and Competitiveness (MINECO),
the European Regional Development Fund (ERDF) under Grant ENE2015-64087-C2-2-R and the Spanish Ministry
of Education (FPU15/01274).
Author Contributions: Marian Liberos, Emilio Figueres and Gabriel Garcerá have been developed the main idea,
the model and the control of the system; Marian Liberos and Raúl González-Medina have been programming the
PSIMTM simulations; Marian Liberos, Emilio Figueres and Gabriel Garcerá have written the paper.
Conflicts of Interest: The authors declare no conflict of interest. The founding sponsors had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the
decision to publish the results.

References
1. Pazheri, F.R.; Othman, M.F.; Malik, N.H. A review on global renewable electricity scenario. Renew. Sustain.
Energy Rev. 2014, 31, 835–845. [CrossRef]
2. Subudhi, B.; Pradhan, R. A Comparative Study on Maximum Power Point Tracking Techniques for
Photovoltaic Power Systems. IEEE Trans. Sustain. Energy 2013, 4, 89–98. [CrossRef]
3. Borrega, M.; Marroyo, L.; González, R.; Balda, J.; Agorreta, J.L. Modeling and Control of a Master–Slave PV
Inverter With N-Paralleled Inverters and Three-Phase Three-Limb Inductors. IEEE Trans. Power Electron.
2013, 28, 2842–2855. [CrossRef]
4. Duman, T.; Marti, S.; Moonem, M.A.; Kader, A.A.R.A.; Krishnaswami, H. A Modular Multilevel Converter
with Power Mismatch Control for Grid-Connected Photovoltaic Systems. Energies 2017, 10, 698. [CrossRef]
5. Araujo, S.V.; Zacharias, P.; Mallwitz, R. Highly Efficient Single-Phase Transformerless Inverters for
Grid-Connected Photovoltaic Systems. IEEE Trans. Ind. Electron. 2009, 57, 3118–3128. [CrossRef]
6. Satcon, Pv Inverters. PowerGate Plus 500 kW. Available online: http://www.satcon.com (accessed on
18 August 2017).
7. Agorreta, J.L.; Borrega, M.; López, J.; Marroyo, L. Modeling and Control of N-Paralleled Grid-Connected
Inverters With LCL Filter Coupled Due to Grid Impedance in PV Plants. IEEE Trans. Power Electron. 2010, 26,
770–785. [CrossRef]
8. Power Electronics. Available online: http://www.power-electronics.com (accessed on 18 August 2017).
9. ABB, ABB Central Inverters. PVS980—1818 to 2091 kVA. Available online: http://new.abb.com (accessed on
18 August 2017).
10. Infineon, Central Inverter Solutions. Available online: https://www.infineon.com/cms/en/applications/
solar-energy-systems/central-inverter-solutions/ (accessed on 18 August 2017).
11. Xiao, H.F.; Xie, X.J.; Chen, Y.; Huang, R.H. An optimized transformerless Photovoltaic Grid-Connected
Inverter. IEEE Trans. Ind. Electron. 2011, 58, 1887–1895. [CrossRef]
12. Ye, Z.; Boroyevich, D.; Lee, F. Modeling and control of zero-sequence current in parallel multi-phase
converters. In Proceedings of the Power Electronics Specialists Conference, Galway, Ireland, 23–23 June 2000.
13. Mazumder, S.M. A novel discrete control strategy for independent stabilization of parallel three-phase boost
converters by combining space-vector modulation with variable-structure control. IEEE Trans. Power Electron.
2003, 18, 1070–1083. [CrossRef]
14. Pan, C.T.; Liao, Y.H. Modeling and Control of Circulating Currents for Parallel Three-Phase Boost Rectifiers
With Different Load Sharing. IEEE Trans. Ind. Electron. 2008, 55, 2776–2785.
15. Ogasawara, S.; Takagaki, J.; Akagi, H.; Nabae, A. A novel control scheme of a parallel current-controlled
PWM inverter. IEEE Trans. Ind. Appl. 1992, 28, 1023–1030. [CrossRef]
16. Ye, Z.H.; Boroyevich, D.; Choi, J.Y.; Lee, F.C. Control of circulating current in two parallel three-phase boost
rectifiers. IEEE Trans. Power Electron. 2002, 2, 609–615.
Energies 2017, 10, 1242 25 of 25

17. Figueres, E.; Garcera, G.; Sandia, J.; Gonzalez-Espin, F.; Rubio, J.C. Sensitivity Study of the Dynamics of
Three-Phase Photovoltaic Inverters With an LCL Grid Filter. IEEE Trans. Ind. Electron. 2009, 56, 56–706.
[CrossRef]
18. Figueres, E.; Garcera, G.; Sandia, J.; Gonzalez-Espin, F.; Calvo, J.; Vales, M. Dynamic Analysis of Three-phase
Photovoltaic Inverters with a High Order Grid Filter. In Proceedings of the IEEE International Symposium
on Industrial Electronics, Vigo, Spain, 4–7 June 2007.
19. Mohd, A.; Ortjohann, E.; Hamsic, N.; Sinsukthavorn, W.; Lingemann, M.; Schmelter, A.; Morton, D. Control
strategy and space vector modulation for three-leg four-wire voltage source inverters under unbalanced
load conditions. IET Power Electron. 2010, 3, 323–333. [CrossRef]
20. Albatran, S.; Fu, Y.; Albanna, A.; Schrader, R.; Mazzola, M. Hybrid 2D-3D Space Vector Modulation Voltage
Control Algorithm for Three Phase Inverters. IEEE Trans. Sustain. Energy 2013, 4, 734–744. [CrossRef]

© 2017 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/).