energies

Article
A Systematic Controller Design for a Grid-Connected
Inverter with LCL Filter Using a Discrete-Time
Integral State Feedback Control and State Observer
Seung-Jin Yoon, Ngoc Bao Lai ID
and Kyeong-Hwa Kim * ID

Department of Electrical and Information Engineering, Seoul National University of Science and Technology,
232 Gongneung-ro, Nowon-gu, Seoul 01811, Korea; tmdwls3233@naver.com (S.-J.Y.);
laingocbao@gmail.com (N.B.L.)
* Correspondence: k2h1@seoultech.ac.kr; Tel.: +82-2-970-6406; Fax: +82-2-978-2754

Received: 16 January 2018; Accepted: 11 February 2018; Published: 16 February 2018

Abstract: Inductive-capacitive-inductive (LCL)-type filters are currently preferred as a replacement

for L-type filters in distributed generation (DG) power systems, due to their superior harmonic
attenuation capability. However, the third-order dynamics introduced by LCL filters pose a challenge
to design a satisfactory controller for such a system. Conventionally, an LCL-filtered grid-connected
inverter can be effectively controlled by using a full-state feedback control. However, this control
approach requires the measurement of all system state variables, which brings about more complexity
for the inverter system. To address this issue, this paper presents a systematic procedure to design
an observer-based integral state feedback control for a LCL-filtered grid-connected inverter in the
discrete-time domain. The proposed control scheme consists of an integral state feedback controller
and a full-state observer which uses the control input, grid-side currents, and grid voltages to
predict all the system state variables. Therefore, only the grid-side current sensors and grid voltage
sensors are required to implement the proposed control scheme. Due to the discrete-time integrator
incorporated in the state feedback controller, the proposed control scheme ensures both the reference
tracking and disturbance rejection performance of the inverter system in a practical and simple way.
As a result, superior control performance can be achieved by using the reduced number of sensors,
which significantly reduces the cost and complexity of the LCL-filtered grid-connected inverter
system in DG applications. To verify the practical usefulness of the proposed control scheme, a 2 kW
three-phase prototype grid-connected inverter has been constructed, and the proposed control system
has been implemented based on 32-bit floating-point digital signal processor (DSP) TMS320F28335.
The effectiveness of the proposed scheme is demonstrated through the comprehensive simulation
and experimental results.

Keywords: discrete-time model; DSP TMS320F28335; grid-connected inverter; integral state feedback
control; LCL filter; state observer

1. Introduction
Renewable energy sources, such as wind and solar, are becoming reliable alternatives to fossil fuel.
In order to integrate these renewable energies into the utility grid, distributed generation (DG) power
systems have drawn much attention from academia, because of their effectiveness and reliability [1,2].
Due to the strong requirement of the high energy efficiency and the flexibility to operate in both the
grid-connected and island modes, power electronic converters are typically used in DG application
systems. DG systems should be able to inject high quality current into the grid to fulfill the grid
interconnection standard [3], which is often ensured by the current controller of a grid-connected
inverter. In particular, the control system of a grid-connected inverter should be implemented in an

Energies 2018, 11, 437; doi:10.3390/en11020437 www.mdpi.com/journal/energies

Energies 2018, 11, 437 2 of 20

economical manner, and its design process should be systematic and straightforward, for easiness to
accept it in DG power systems.
The key aspect of utilizing a grid-connected inverter is to meet the interface standards as stated
in [4,5], and to minimize the implementation cost of the DG systems. Formerly, the grid-connected
inverter employs only L-type filter as a mean of suppressing current harmonics injected into the
grid [6,7]. In recent years, however, LCL-type filters are more attractive than the L-type filter, due to
the smaller physical size and better harmonic attenuation capability in spite of the control complexity.
In order to preserve the power quality, the total harmonic distortion (THD) in injected current needs to
meet certain standards, such as IEEE-519 or IEC 61000-3-2, which have been enacted for the purpose of
satisfying the gird integration codes [8].
Conventionally, to control grid-connected inverters, the linear controllers, such as the
proportional-integral (PI) control, have been employed in the synchronous reference frame, due
to their simplicity and stability [9]. However, the conventional PI decoupling controller does not give
proper performance in the inverter system having LCL filters. Even though this control scheme can
ensure a moderate quality of injected current into the utility grid, the choice of controller gains often
requires a process of trial and error [10].
As other approaches, the proportional-resonant (PR) controller, the predictive controller [11],
and the repetitive feedback controller [12–14] have been proposed in literature. The PR controllers have
been widely used for grid-connected inverters, due to their capability to reject individual harmonics
by introducing an infinite gain at a selected resonance frequency [15–20]. The implementation and
evaluation of a typical PR current controller in an inverter system was reported in [17,18]. In order
to consider the parameter variations, a direct discrete-time pole placement strategy was proposed
in [19] for the parameter design of a PR controller. In addition to the PR controller, an efficient hybrid
damper for an LCL- or LLCL-filter has been presented for a single-phase grid-tied inverter, by using the
capacitor current feedback to avoid the potential instability of system [20]. By implementing multiple
PR controllers, selected higher-order harmonics can be effectively rejected. However, the resonant
controllers usually increase the complexity and calculation burden of the system.
On the other hand, predictive controllers give a satisfactory performance under the assumption
that the system parameters are exactly known. This controller mimics the deadbeat control in the
synchronous dq-reference frame, and has the advantage of simple implementation. In order to minimize
the resonant behavior during the grid connection, a predictive controller has been designed by using
the discrete-time model of inverter system in the natural reference frame [11].
A repetitive feedback control, which is based on the concept of iterative learning control, has been
widely applied for many practical industrial systems. The research work in [14] presented a repetitive
control scheme, in which the inverter-side currents are regulated by the PI controllers, and the grid-side
currents are controlled by a separate repetitive controller. Due to the use of the repetitive controller,
considerable improvement under harmonically distorted grid voltage can be observed.
To further improve the robustness of inverter system, an H-infinity repetitive current controller
has been presented [21]. Since the parameter variations were taken into account in designing the
controller by H-infinity approach, the robustness against the parameter variations of inverter system
can be improved. Aside from the repetitive control scheme, a neural network vector controller has
been presented for a single-phase grid-connected inverter [22]. In this work, the controller was trained
by the Levenberg-Marquardt algorithm, to eliminate the influence of coupling terms. However, it is
worth mentioning that the use of learning mechanism in these controllers often gives slow dynamic
response of the closed-loop system.
This paper proposes a systematic controller design methodology for a grid-connected inverter with
LCL filter using a discrete-time integral state feedback control and state observer schemes. Generally,
to implement the integral state feedback controller, the measurement of all system states should be
required, which increases the cost and complexity of system, due to the necessity of additional sensors.
To overcome such a limitation, this paper presents a systematic design procedure for a LCL-filtered
Energies 2018, 11, 437 3 of 20

grid-connected inverter in a discrete-time domain, by using the estimated states based on full-state
observer. The proposed control scheme consists of an integral state feedback controller and a full-state
observer which uses the control input, grid-side currents, and grid voltages to predict all the system
state variables. The proposed control scheme ensures both the reference tracking and disturbance
rejection performance of the inverter system by using only the measurement of the grid-side currents
and grid voltages. This significantly reduces the cost and complexity of the LCL-filtered grid-connected
inverter system in DG applications. In addition, the discrete-time integrator is also incorporated in
the state feedback controller, which ensures an asymptotic reference tracking performance in practical
and simple way. For the purpose of verifying the practical usefulness of the proposed control scheme
in DG applications, the whole control algorithm has been implemented using the software of 32-bit
floating-point digital signal processor (DSP) TMS320F28335 to control a 2 kW three-phase prototype
grid-connected inverter [23]. The feasibility and effectiveness of the proposed scheme are demonstrated
through the comprehensive simulations and experiments. In addition to providing a systematic
design methodology, the proposed control scheme can be a very practical approach of designing a
grid-connected inverter in DG applications by virtue of its simplicity and low cost.

2. Discrete-Time Model of a Grid-Connected Inverter with LCL Filter

2.1. Modeling of a Grid-Connected Inverter with LCL Filter

Since the state feedback control is a model-based approach, the precise system model and
the discretization process are the key factors to design the controller successfully. To obtain the
system model in the synchronous reference frame, all three-phase variables are transformed into the
synchronous reference frame using Park’s transformation, as follows:
h iT h iT
q
i2 i2d = T i2a i2b i2c (1)

h iT h iT
q
i1 i1d = T i1a i1b i1c (2)
h iT h iT
q
vc vdc = T vca vbc vcc (3)

where the subscript “qd” denotes the variables on the synchronous reference frame, the subscript
h iT
q
“abc” denotes the variables in three-phase coordinate, i2 i2d is the grid-side current vector,
h iT h iT
q q
i1 i1d is the inverter-side current vector, vc vdc is the capacitor voltage vector, and T is the
transformation matrix, expressed as
"
#
cos θ − 23 π cos θ + 23 π


2 cos θ
T= (4)
sin θ − 32 π sin θ + 32 π



3 sin θ

Figure 1 shows a three-phase grid-connected inverter with an LCL filter, in which VDC denotes
the DC-link voltage, R1 , R2 , L1 , and L2 are the filter resistances and filter inductances, respectively,
and Cf is the filter capacitance. From Figure 1, the inverter system can be mathematically expressed in
the synchronous reference frame as
.
q R2 q 1 q 1
i2 = − i − ωi2d + vc − eq (5)
L2 2 L2 L2
.
R2 d q 1 1
i2d = − i2 + ωi2 + vdc − ed (6)
L2 L2 L2
.
q R1 q 1 q 1 q
i1 = − i − ωi1d − vc + vi (7)
L1 1 L1 L1
Energies 2018, 11, 437 4 of 20
Energies 2018, 11, x FOR PEER REVIEW 4 of 19

.
d R1 d q 1 1 d1 1 d
i −1 − L +
=1−= − L +i1 + ωi v c + vi
L1
(8)
(8)
1 1
.
q d 1 1 q1 1 q
v=c −= −ωv
−c − C +i2 + C i1 (9)
(9)
f f
.
q 1 1 d1 1 d
vdc = ωv
= −c − C +i2 + C i1 (10)
(10)
f f
whereω denotes
where denotesthe
the angular
angular frequency
frequency of the
of the gridgrid voltage,
voltage, e is the is thevoltage,
grid grid voltage,
and vi isand is the
the inverter
invertervoltage.
output output voltage.

PWM inverter i1 R1 L1 R2 L2 i2
•
+
VDC +
_ vc Cf Grid
_

Figure 1. Configuration of the simplified equivalent circuit of three-phase grid-connected inverter

Figure 1. Configuration of the simplified equivalent circuit of three-phase grid-connected inverter with
with LCL filter. PWM: pulse width modulation
LCL filter. PWM: pulse width modulation

From (5) to (10), the state-space model of the inverter system can be obtained as
From (5) to (10), the state-space model of the inverter system can be obtained as
( )= ( )+ ( )+ ( ) (11)
.
x(t) = Ax(t) + Bu(t) + De(t) (11)
( )= ( ) (12)
y(t) = Cx(t) (12)
where = is the system state vector, = is the system input vector,
[ iq i]d iqisidthe
iT iT
and x= = q grid voltage vector, state
and the system u matrices
= vi vid , is ,the, and
system are expressed
h h
d q
where 2 2 1 1 vc vc is the system vector, input vector,
as h iT
and e = eq ed is the grid voltage vector, − and the0system matrices A, B, C, and D are expressed as
− / 0 1/ 0
− / 0 0 0 1/
−R−2 /L/2 −ω0 − 0 −1/
 
=
00

1/L2 0 0
 ω0 −−R2 /L
/
2 0 00 00 −1/1/L2 
0 1/ 0
 − R−1/ −ω −01/L1 − 0
 
1 /L1 0 0 
A= −1/ 0 1/
−01/L1 
 
00 − R1 /L1 0 0

 ω
 
0 −1/C 0 f 0
−1/
1/C f
0
0 0 −ω 

0 00 −1/C0f 0−1/ 1/C f ω 0
1/ 0 , = 0 0 , C = 1 0 0 0 0 0 .
=
0 0 1/ −01/L2 0 0 0 1 0 0 0 0

0
 0 0 0 0  00 0 − 1/L 
2 
0 0 0  00 0 0
   " #
 1/L  1 0 0 0 0 0
1
B= , D =  , C = .
   
 0 1/L1   0 0  0 1 0 0 0 0
2.2. Discrete-Time 0 Model 0
   
   0 0 
0
For the purpose of a 0digital implementation,
0 0
the continuous-time model of the inverter system
can be discretized as
2.2. Discrete-Time Model
( + 1) = ( )+ ( )+ ( ) (13)
For the purpose of a digital implementation, the continuous-time model of the inverter system
can be discretized as
( )= ( ) (14)
x(k + 1) = Ad x(k ) + Bd u(k) + Dd e(k) (13)
where the matrices , , , and can be calculated with the sampling period as
y ( k ) = Cd x ( k ) (14)
= = + + +⋯
1! 2!
Energies 2018, 11, 437 5 of 20

where the matrices Ad , Bd , Cd , and Dd can be calculated with the sampling period Ts as
Energies 2018, 11, x FOR PEER REVIEW 5 of 19
ATs ATs A2 Ts2
Ad = e = I+ + +···
1! 2!
= ( − ) , =
Bd = A−1 (Ad − I)B, Cd = C

Dd ==A−1 ((Ad −
− I))D.
.

3. Design
3. Design of
of aa Discrete-Time
Discrete-Time Integral
Integral State
State Feedback
Feedback Control
Control with
with State
State Observer
Observer
Figure 22 shows
Figure shows thethe block
block diagram
diagram of of the
the proposed
proposed control
control scheme
scheme forfor aa three-phase inverter
three-phase inverter
connected to the grid through LCL filters, where the symbol “ˆ” denotes the
connected to the grid through LCL filters, where the symbol “^” denotes the estimated quantities. estimated quantities.
The proposed
The proposedcontrol
controlscheme
scheme basically consists
basically of anofintegral
consists state feedback
an integral controller,
state feedback and a full-state
controller, and a
full-state observer with the measurement of only the grid-side currents and grid voltages. grid
observer with the measurement of only the grid-side currents and grid voltages. The measured The
voltages are
measured alsovoltages
grid employed areinalso
the employed
phase lockin loop
the(PLL),
phaseto generate
lock the phase
loop (PLL), angle of the
to generate the phase
grid voltage.
angle
The
of full-state
the observer
grid voltage. Theisfull-state
constructed by using
observer an LCL-filtered
is constructed inverter
by using model to estimate
an LCL-filtered inverterthe system
model to
states x from the control input u and the system output y. Then, the estimated
estimate the system states from the control input and the system output y. Then, the estimated states are used for the
integral
states arestate
usedfeedback controller
for the integral to track
state feedbackthe reference
controllercurrents.
to track the reference currents.

Figure 2. Proposed
Figure 2. Proposed discrete-time
discrete-time integral
integral state
state feedback
feedback control
control scheme
scheme with
with state
state observer
observer for
for
LCL-filtered three-phase grid-connected inverter. PLL: phase lock loop;
LCL-filtered three-phase grid-connected inverter. PLL: phase lock loop.

The system in (13) and (14) is controllable if, and only if, the controllability matrix expressed
The system in (13) and (14) is controllable if, and only if, the controllability matrix Pc expressed as
as
i
= [[Bd Ad Bd A2d Bd . .…
Pc = . Ad n−1 Bd] (15)
(15)
has full rank. Since the rank of a matrix is 6, the full-state vector is completely controllable.
Similarly, the observable condition of the system in (13) and (14) can be investigated through the
observability matrix as follows:

= . (16)
⁞

Since the rank of Po is 6, the full-state vector is completely observable.

One of the key tasks of the controller is to stabilize the inverter system, which can be
accomplished by using a state feedback regulator. The control input of the state feedback regulator
can be expressed as
Energies 2018, 11, 437 6 of 20

has full rank. Since the rank of a matrix Pc is 6, the full-state vector is completely controllable. Similarly,
the observable condition of the system in (13) and (14) can be investigated through the observability
matrix as follows:  
Cd
 Cd A d 
 
Po =  .. . (16)
.
 
 
Cd A d n − 1
Since the rank of Po is 6, the full-state vector is completely observable.
One of the key tasks of the controller is to stabilize the inverter system, which can be accomplished
by using a state feedback regulator. The control input of the state feedback regulator can be expressed as

u(k) = −Kx(k) (17)

where K denotes the gain matrix of the state feedback regulator.

Generally, the state feedback regulator is unable to track the reference currents in the presence
of the grid voltage disturbances and the variations of inverter parameters. Therefore, an integral
control is introduced in this paper to assure the stable and robust operation of the inverter system. To
construct the integral state feedback control, the discrepancy of the measured grid-side currents from
their reference values can be given as

xe (t) = r(t) − y(t) = r − Cd x(t) (18)

h iT
q∗
where r = i2 i2d∗ is the reference current vector and xe is the current error vector. From (18), an
integral action can be formulated in a state-space representation as follows:
.
xi (t) = 0xi (t) + Ixe (t). (19)

The discrete-time representation of (19) can be obtained by using the discretization procedure
presented in the previous section as follows:

xi (k + 1) = xi (k) + Ts r − Ts Cd x(k). (20)

The system in (20) can be augmented in the inverter system in (13) and (14) as
" # " #" # " # " # " #
x( k + 1) Ad 0 x( k ) Bd Dd 0
= + u( k ) + e( k ) + Ts r(k) (21)
xi (k + 1) − Ts Cd I xi (k) 0 0 I
" #
x( k )
y ( k ) = [ Cd 0 ] . (22)
xi (k)

It can be observed from (21) and (22) that the asymptotic reference tracking is ensured as long
as the augmented system is stabilized, i.e., the state xi is derived to zero. For this purpose, a state
feedback control is employed for this augmented system as follows:
" #
h i x( k )
u( k ) = − K Ki (23)
xi (k)
Energies 2018, 11, 437 7 of 20

h i
where K Ki is the gain vector, which can be systematically evaluated by using the pole placement
method. Substituting (23) in (21), the state equation for the closed-loop system can be derived as
follows:
" # " # " # !" # " # " #
x( k + 1) Ad 0 Bd h i x( k ) Dd 0
= − K Ki + e( k ) + Ts r(k). (24)
xi (k + 1) − Ts Cd I 0 xi ( k ) 0 I

The closed-loop system in (24) is stable if, and only if, all the eigenvalues of the equation as
" # " # !!
Ad 0 Bd h i
det λI − − K Ki =0 (25)
− Ts Cd I 0

lie within the unit circle. It is worth mentioning that the feedback controller in (23) requires the
measurement for all the system states. However, this requirement often makes the system design
costly and complicated. In order to avoid this limitation by implementing the integral state feedback
controller by using the measurement of only the grid voltages and grid-side currents, a full-state
observer is employed for the given system.
From the inverter system model in (13) and (14), the state equation of the full-state observer can
be expressed as
x̂(k + 1) = Ad x̂(k) + Bd u(k) + Dd e(k ) + G(y(k ) − Cd x̂(k)) (26)

where the symbol “ˆ” denotes the estimated variables, and G is the observer gain matrix. If the
x = x(k) − x̂(k ), the error dynamics of the observer can be obtained by
estimation error is defined as e
subtracting (26) from (13) as follows:

x(k + 1) = (Ad − GCd )e

e x( k ). (27)

For the state observer to be stable, all the eigenvalues of the equation

det(λI − (Ad − GCd )) = 0. (28)

should lie within the unit circle. Then, the state feedback control input can be constructed by using the
estimated state variables as " #
h i x̂(k)
u ( k ) = − K Ki . (29)
xˆi (k )

The two gain vectors in (29) can be systematically evaluated by using the pole placement method.
The selection of the observer poles requires a compromise between the rapidity of response and the
sensitivity to measurement noise. The observer poles of −α ± jβ in the continuous-time domain can
be transferred to the discrete-time domain through the transformation of z = e Ts s as follows:

ζ ± jη = e−αTs (cos βTs ± j sin βTs ). (30)

Thus, if the observer poles in the continuous-time domain are determined, the pole locations
in z-domain, and thus, the observer gains can be easily obtained by the pole placement technique.
Figure 3 illustrates the control block diagram of the proposed integral state feedback control scheme in
the discrete-time domain.
 ζ = (cos sin ). (30)
Thus, if the observer poles in the continuous-time domain are determined, the pole locations in
z-domain, and thus, the observer gains can be easily obtained by the pole placement technique.
Figure
Energies 3 illustrates
2018, 11, 437 the control block diagram of the proposed integral state feedback control scheme
8 of 20
in the discrete-time domain.

Figure 3. Control
Control block
block diagram
diagram of
of the
the proposed
proposed integral
integral state
state feedback
feedback control
control scheme
scheme in
in the
discrete-time domain.

4.
4. Simulation
Simulation Results
Results
In ordertotoprove
In order provethethe feasibility
feasibility of the
of the proposed
proposed control
control scheme,
scheme, the simulations
the simulations have
have been been
carried
carried out
out for an for an LCL-filtered
LCL-filtered three-phase three-phase grid-connected
grid-connected inverter
inverter using using
the PSIM the PSIM
software. Thesoftware. The
configuration
configuration of the inverter system and the proposed control scheme are depicted in
of the inverter system and the proposed control scheme are depicted in Figure 3. Also, the systemFigure 3. Also,
the system parameters
parameters are summarizedare summarized
in Table 1. in Table 1.

Table
Table 1.
1. System
System parameters
parameters of
of aa grid-connected inverter.
grid-connected inverter.

Parameters Symbol Value Units

Parameters Symbol Value Units
Grid voltage E 180 V
Grid voltage E 180 V
Grid frequency F 60 Hz
Grid frequency F 60 Hz
DC-link
DC-link voltage voltage VDC V DC
420 420 V V
Inverter-side filter
Inverter-side filter inductance L 1 0.9 mH
L1 0.9 mH
inductance
Grid-side filter inductance L2 1.7 mH
Grid-side filter
Filter resistance L2 R1 , R 1.7
2 0.5 mH Ω
inductance
Filter capacitor R , R
Filter resistance
Cf 0.5 5 Ω
1 2
Filter capacitor Cf 5 µF

Figure 4 shows the simulation results for the inverter states and estimated states, using the
proposed integral state feedback control scheme with state observer at the synchronous reference
frame (SRF). To locate the observer closed-loop poles at the desired locations, the observer gains are
chosen by using Matlab “place” function with inverter parameters. As can be observed from Figure 4,
the estimated states instantly converge to the actual ones, regardless of oscillations during initial
transient periods. In this simulation, the phase lock loop (PLL) also starts its operation at t = 0, to
determine the phase angle of grid voltage. The large initial oscillation in the estimated states partly
comes from the fact that the PLL requires sufficient time to be stabilized. After the PLL reaches the
steady state, the estimated states will converge to the actual ones, which confirms a fast and stable
operation of the observer. These estimated states are employed in (24) to constitute the integral state
feedback control scheme in a discrete-time domain.
transient periods. In this simulation, the phase lock loop (PLL) also starts its operation at t = 0, to
determine the phase angle of grid voltage. The large initial oscillation in the estimated states partly
comes from the fact that the PLL requires sufficient time to be stabilized. After the PLL reaches the
steady state, the estimated states will converge to the actual ones, which confirms a fast and stable
operation of11,
Energies 2018, the437observer. These estimated states are employed in (24) to constitute the integral 9state
of 20
feedback control scheme in a discrete-time domain.

(a)

(b)

(c)
Figure 4. Simulation results for the inverter states and estimated states using the proposed
Figure 4. Simulation results for the inverter states and estimated states using the proposed discrete-time
discrete-time integral state feedback control scheme with state observer. (a) Waveforms of grid-side
integral state feedback control scheme with state observer. (a) Waveforms of grid-side currents and
currents and estimated states at the synchronous reference frame (SRF); (b) Waveforms of
estimated states at the synchronous reference frame (SRF); (b) Waveforms of inverter-side currents and
inverter-side currents and estimated states at the SRF; (c) Waveforms of capacitor voltages and
estimated states at the SRF; (c) Waveforms of capacitor voltages and estimated states at the SRF.
estimated states at the SRF.

Figure 55 shows
Figure shows the
the simulation results for
simulation results for the
the proposed
proposed integral
integral state
state feedback control scheme
feedback control scheme
with state observer. Similarly, to place the closed-loop poles of the integral state feedback
with state observer. Similarly, to place the closed-loop poles of the integral state feedback controlcontrol
scheme at the desired locations, the control gain matrices are chosen by using
scheme at the desired locations, the control gain matrices are chosen by using Matlab “place” function
Matlab “place”
q
with inverter LCL parameters. As is shown in Figure 5a, the grid-side currents i2 and i2d well track the
reference currents at the SRF with satisfactory dynamic performance, without showing any steady-state
errors or fluctuation. Even though the transient response of grid-side currents shows large settling
time and some oscillations during start-up period, this is caused by the required time for the PLL to be
stabilized. If the PLL reaches the steady state, the actual grid-side q-axis current tracks the reference
within 0.5 ms. Also, Figure 5b,c represent the measured waveforms of the inverter-side currents and
capacitor voltages at the SRF.
 well track the reference currents at the SRF with satisfactory dynamic performance, without
showing any steady-state errors or fluctuation. Even though the transient response of grid-side
currents shows large settling time and some oscillations during start-up period, this is caused by the
required time for the PLL to be stabilized. If the PLL reaches the steady state, the actual grid-side
q-axis current tracks the reference within 0.5 ms. Also, Figure 5b,c represent the measured
Energies 2018, 11, 437 10 of 20
waveforms of the inverter-side currents and capacitor voltages at the SRF.

(a)

(b)

(c)
Figure 5. Simulation results for the proposed discrete-time integral state feedback control scheme
Figure 5. Simulation results for the proposed discrete-time integral state feedback control scheme with
with state observer. (a) Current references and grid-side current responses at the SRF; (b)
state observer. (a) Current references and grid-side current responses at the SRF; (b) Inverter-side
Inverter-side
current current
responses responses
at the SRF; (c)at the SRF; voltage
Capacitor (c) Capacitor voltage
responses responses
at the SRF. at the SRF.

Figure 6 represents the simulation results for three-phase variables using the proposed integral
Figure 6 represents the simulation results for three-phase variables using the proposed integral
state feedback control scheme with state observer. As can be seen from Figure 6a, three-phase
state feedback control scheme with state observer. As can be seen from Figure 6a, three-phase grid-side
grid-side current waveforms are relatively sinusoidal. In fact, the grid-side phase currents have the
current waveforms are relatively sinusoidal. In fact, the grid-side phase currents have the THD level
THD level of 2.57% in this case. Also, Figure 6b,c show actual three-phase inverter-side current
of 2.57% in this case. Also, Figure 6b,c show actual three-phase inverter-side current waveforms and
waveforms and three-phase capacitor voltage waveforms.
three-phase capacitor voltage waveforms.
Figure 7 shows the simulation results for the proposed integral state feedback control scheme
with state observer under the step change in q-axis current reference. In this simulation, the q-axis
reference current undergoes a step change from 7 to 10 A at 0.05 s. It is clearly shown in Figure 7a
that the grid-side currents reach the reference current values with a considerably small time delay.
Since the PLL already reached the steady state during start-up period, the grid-side q-axis current
can effectively track the step change in reference value within 0.5 ms, without showing such a large
oscillation in this case. This indicates the sufficiently fast transient response of the proposed control
scheme. Additionally, the fast transient response of the proposed control scheme can be inferred from
Figure 7b,c.
 Energies 2018, 11, 437 11 of 20
Energies 2018, 11, x FOR PEER REVIEW (a) 10 of 19

(b)
(a)

(b)
(c)
Figure 6. Simulation results for three-phase variables using the proposed discrete-time integral state
feedback control scheme with state observer. (a) Three-phase grid-side current waveforms; (b)
Three-phase inverter-side current waveforms; (c) Three-phase capacitor voltage waveforms.

Figure 7 shows the simulation results for the proposed integral state feedback control scheme
with state observer under the step change in q-axis current reference. In this simulation, the q-axis
reference current undergoes a step change from 7 to 10 A at 0.05 s. It is clearly shown in Figure 7a
that the grid-side currents reach the reference current values with a considerably small time delay.
(c)
Since the PLL already reached the steady state during start-up period, the grid-side q-axis current
Figure 6. Simulation
can effectively results for three-phase variables using the proposed
ms, discrete-time integralsuch
state
Figure 6. track the step
Simulation change
results in reference
for three-phase value
variables within 0.5proposed
using the without showing
discrete-time integral a large
feedback
oscillation control scheme with state observer. (a) Three-phase grid-side current waveforms; (b) control
statein this case.
feedback Thisscheme
control indicates
withthe sufficiently
state observer. (a)fast transient response
Three-phase of the waveforms;
grid-side current proposed
Three-phase
scheme.(b) inverter-side
Additionally, the fastcurrent waveforms;
transient (c) Three-phase
response capacitor control
voltage waveforms.can be inferred
Three-phase inverter-side current waveforms; (c)of the proposed
Three-phase capacitor voltagescheme
waveforms.
from Figure 7b,c.
Figure 7 shows the simulation results for the proposed integral state feedback control scheme
with state observer under the step change in q-axis current reference. In this simulation, the q-axis
reference current undergoes a step change from 7 to 10 A at 0.05 s. It is clearly shown in Figure 7a
that the grid-side currents reach the reference current values with a considerably small time delay.
Since the PLL already reached the steady state during start-up period, the grid-side q-axis current
can effectively track the step change in reference value within 0.5 ms, without showing such a large
oscillation in this case. This indicates the sufficiently fast transient response of the proposed control
scheme. Additionally, the fast transient response of the proposed control scheme can be inferred
from Figure 7b,c.

(a)

Figure 7. Cont.

(a)
Energies 2018, 11, x FOR PEER REVIEW 11 of 19

Energies 2018, 11, 437 12 of 20

Energies 2018, 11, x FOR PEER REVIEW 11 of 19

(b)

(b)

(c)
Figure 7. Simulation results for the proposed discrete-time (c) integral state feedback control scheme
with state
Figureobserver under
7. Simulation the step
results change
for the proposedin q-axis current
discrete-time reference.
integral (a) Current
state feedback references
control scheme and
Figure 7. Simulation results for the proposed discrete-time integral state feedback control scheme with
grid-side
withcurrent responses
state observer under at the
thestep
SRF;change
(b) Inverter-side current
in q-axis current responses
reference. at thereferences
(a) Current SRF; (c) Capacitor
and
state observer under the step change in q-axis current reference. (a) Current references and grid-side
voltage responses
grid-side
current at the
current
responses at SRF.
responses at the SRF; (b) Inverter-side current responses at the SRF; (c)
the SRF; (b) Inverter-side current responses at the SRF; (c) Capacitor voltage
Capacitor
voltage responses at the SRF.
responses at the SRF.
Figures 8 and 9 show the simulation results for the proposed discrete-time integral state
Figures 8 and 9 show the simulation results for the proposed discrete-time integral state
feedback control
Figures 8scheme
andscheme
9 showwith the state observer resultsunder islanding modes. In integral
this operating mode, the
feedback control withsimulation
state observer for theislanding
under proposedmodes.
discrete-time
In this operating statemode,
feedbackthe
capacitor voltages
control scheme are
withcontrolled
state with
observer their
under reference
islanding values
modes. to
In supply
this critical
operating
capacitor voltages are controlled with their reference values to supply critical loads. One of the key
loads.
mode, the One of
capacitor the key
roles voltages
of theofgrid-connected
roles are controlled
the with
grid-connected inverter
their isistotoensure
reference
inverter values not
ensure onlyacritical
to supply
not only astable
stable operation
loads. Oneof
operation of inverter,
of inverter,
the key roles ofbut
but also theaalso a
grid-connected
continuous
continuoussupply inverter is to
of electricity
supply ensure
of electricity not only
totocritical a stable
critical loads. operation
When the
loads. When of inverter, but
thegrid-connected
grid-connectedalso a continuous
inverter
inverter supply
undergoes
undergoes a a
suddenof electricity
operation
sudden to critical
mode
operation modeloads. When
transition, the
transition,the grid-connected
thevoltages
voltages at inverter
at the
the point undergoes
pointofofcommon
commona sudden operation
coupling
coupling mode
oftenoften
exhibit exhibit
transition,
transient the voltages
behavior. at the point
However, of common
such coupling often exhibit transient behavior. However,
transient behavior. However, such a a transient
transient behavior
behavior isisnotnotdesirabledesirable to critical loads.loads.
to critical
such a transient
Furthermore, behavior
the is not desirable
grid-connected inverterto iscritical
often loads.
required Furthermore,
totooperate the grid-connected
continuously, eveninverter
during
Furthermore,
is often
the grid-connected
required to operate
inverter
continuously,
is often
even
required
during islanding
operate
modes,
continuously,
to supply critical
even
loads.
during
islanding
islanding modes,
modes, to to supply
supply critical
critical loads. The proposed control scheme has a benefit under this
The proposed
operation control
mode, sincescheme has aloads.
the capacitor benefit The
voltagesunder proposed
are this
control
operation
continually
scheme
mode,
estimated since
andthe
hascapacitor
a benefit under this
voltages
can be controlled by
operation
are mode, since
continually the capacitor
estimated and can be voltages
controlled are
by continually
using these estimated
estimated andin can
values, case be
of controlled
need. It is by
using these estimated values, in case of need. It is confirmed from Figures 8 and 9 that the proposed
usingconfirmed
these estimated
control scheme values,
from works
Figures 8 and
well, in 9case
even that of
theneed.
during It is control
proposed
islanding confirmed
modes scheme from
to supply Figures
works well,
critical 8even
loads. andduring
9 thatislanding
the proposed
modes to supply critical loads.
control scheme works well, even during islanding modes to supply critical loads.

(a)

(a)

Figure 8. Cont.

(b)

(b)
Energies 2018, 11, 437 13 of 20

(a)

Energies 2018, 11, x FOR PEER REVIEW 12 of 19

(b)

(c)
Figure 8. Simulation results for the proposed control scheme under islanding modes at the SRF. (a)
Figure 8. Simulation results for the proposed control scheme under islanding modes at the SRF.
Capacitor voltage responses; (b) Current responses in critical loads; (c) Inverter-side current
(a) Capacitor voltage responses; (b) Current responses in critical loads; (c) Inverter-side current responses.
responses.
responses.

(a)

(b)

(c)
Figure 9. Simulation results for the proposed control scheme under islanding modes at the natural
Figure 9. Simulation results for the proposed control scheme under islanding modes at the natural
reference frame. (a) Capacitor voltage responses; (b) Current responses in critical loads; (c)
reference frame. (a)
Inverter-side Capacitor
current voltage responses; (b) Current responses in critical loads; (c) Inverter-side
responses.
current responses.
5. Experimental Results
For the experimental validation of the proposed control scheme, the experimental setup has
been constructed, which consists of a DSP-based controller, a three-phase inverter connected to the
grid through an LCL filter, a magnetic contactor for grid connecting operation, and an AC power
source to imitate three-phase grid voltages. The control algorithm is performed by a 32-bit
floating-point DSP TMS320F28335 [23], to control a 2 kW laboratory prototype inverter. The
Energies 2018, 11, 437 14 of 20

5. Experimental Results
For the experimental validation of the proposed control scheme, the experimental setup has been
constructed, which consists of a DSP-based controller, a three-phase inverter connected to the grid
through an LCL filter, a magnetic contactor for grid connecting operation, and an AC power source
to imitate three-phase grid voltages. The control algorithm is performed by a 32-bit floating-point
Energies 2018, 11, x FOR
DSP TMS320F28335 [23],PEER
to REVIEW
control a 2 kW laboratory prototype inverter. The configuration 13 of 19 of the

entire experimental system is illustrated in Figure 10a. An intelligent power module is employed for
configuration of the entire experimental system is illustrated in Figure 10a. An intelligent power
three-phase
module is employed for inverter,
grid-connected and
three-phase the switching
grid-connected frequency
inverter, and the (sampling frequency)
switching frequency is chosen as
(sampling
10 kHz. frequency)
Figure 10b shows the photograph of the experimental test setup.
is chosen as 10 kHz. Figure 10b shows the photograph of the experimental test setup.

(a)

(b)
Figure 10. Configuration of the experimental system. (a) Block diagram of the overall system; (b)
Figure 10. Configuration of the experimental system. (a) Block diagram of the overall system;
Photograph of the experimental setup. DSP: digital signal processor.
(b) Photograph of the experimental setup. DSP: digital signal processor.
Figure 11 shows the experimental results for the inverter system using the proposed
discrete-time
Figure 11 shows integral state feedbackresults
the experimental controlfor
scheme with state
the inverter observer.
system usingIt can
the be observeddiscrete-time
proposed from
q-axis and three-phase grid-side currents in Figure 11a that the proposed control scheme provides
integral state feedback control scheme with state observer. It can be observed from q-axis and
stable and reliable operation for the inverter system with considerably sinusoidal responses of
three-phase grid-side
grid-side currentsStable
phase currents. in Figure 11a of
operation that
the the
stateproposed control
observer can scheme
be also provides
observed stable and
from Figure
reliable 11b,c,
operation
in which inverter-side three-phase currents, ̂ , ̂ , and ̂ , are calculated from the estimated phase
for the inverter system with considerably sinusoidal responses of grid-side
currents. Stablê operation
currents and ̂ , andof the state observer
three-phase capacitorcan be also observed
voltages, , , and from Figure
, from 11b,c, in which
the estimated
capacitor voltages and , respectively.
Energies 2018, 11, 437 15 of 20

q
inverter-side three-phase currents, î1a , î1b , and î1c , are calculated from the estimated currents î1 and
q
î1d , and three-phase capacitor voltages, v̂ca , v̂bc , and v̂cc , from the estimated capacitor voltages v̂c and
v̂dc , respectively.
Energies 2018, 11, x FOR PEER REVIEW 14 of 19

(a)

(b)

(c)
Figure 11. Experimental results for the proposed discrete-time integral state feedback control scheme
Figure 11. Experimental results for the proposed discrete-time integral state feedback control scheme
with state observer. (a) Grid-side currents; (b) Estimated inverter-side currents; (c) Estimated
with state observer.
capacitor voltages.(a) Grid-side currents; (b) Estimated inverter-side currents; (c) Estimated
capacitor voltages.
Energies 2018, 11, 437 16 of 20
Energies 2018, 11, x FOR PEER REVIEW 15 of 19

Figure 12
Figure 12 shows
shows the
the experimental
experimental results
results for
for the
the grid-side
grid-side current
current responses
responses ofof the
the proposed
proposed
control scheme, which proves that the experimental results are well consistent with the simulations
control scheme, which proves that the experimental results are well consistent with the simulations in
Figures
in 4a 4a
Figures andand
5a. 5a.
These results
These confirm
results a good
confirm performance
a good of theofproposed
performance controlcontrol
the proposed schemescheme
at steady
at
state in practical realization.
steady state in practical realization.

(a)

(b)
Figure 12. Experimental results for the proposed discrete-time integral state feedback control scheme
Figure 12. Experimental results for the proposed discrete-time integral state feedback control scheme
with state observer. (a) Measured and estimated grid-side currents at the SRF; (b) Grid-side reference
with state observer. (a) Measured and estimated grid-side currents at the SRF; (b) Grid-side reference
currents and measured grid-side currents at the SRF.
currents and measured grid-side currents at the SRF.

For the purpose of validating the proposed control scheme during transient state, Figure 13
showsFor theexperimental
the purpose of validating
results ofthe
theproposed
proposedcontrol scheme
control scheme during
under transient
the stepstate, Figure
change 13 shows
in grid-side
the experimental results of the proposed control scheme under the step change in grid-side
q-axis current reference, from 4 to 6 A. As is shown, the actual grid-side q-axis current tracks q-axis
the
current reference, from 4 to 6 A. As is shown, the actual grid-side q-axis current tracks the reference
reference within 0.5 ms. It is clearly confirmed from this result that the grid-side current reaches the
within 0.5value
reference ms. Itwith
is clearly confirmed
considerably from
small this
time resultwhich
delay, that the grid-side
accords well current
with thereaches the reference
simulation result in
value with
Figure 7. considerably small time delay, which accords well with the simulation result in Figure 7.
Energies 2018, 11, 437 17 of 20
Energies 2018, 11, x FOR PEER REVIEW 16 of 19
Energies 2018, 11, x FOR PEER REVIEW 16 of 19

Figure 13. Experimental results of the proposed discrete-time integral state feedback control scheme
Figure 13. Experimental results of the proposed discrete-time integral state feedback control scheme
with state
Figure 13. observer underresults
Experimental a stepof
change in grid-side
the proposed q-axis current
discrete-time reference.
integral state feedback control scheme
with state observer under a step change in grid-side q-axis current reference.
with state observer under a step change in grid-side q-axis current reference.
To assess the performance of the state observer, Figure 14 shows the experimental results for
To assess
To assessand
the measured theestimated
the performance
performance ofofthe
waveformsthestate
state
of observer,
a-phase Figure
grid-side
observer, Figure 14 14
currentshows
andthe
shows the experimental
inverter-side current.
experimental results
From
results for
for
this the measured
figure, it can and
be estimated
observed waveforms
that the of a-phase
estimated grid-side
states well current
predict theand inverter-side
actual
the measured and estimated waveforms of a-phase grid-side current and inverter-side current. Fromstates. current.
Also, it is
From
worth
this this figure,
mentioning
figure, it can
it can bethat be observed
the estimated
observed that the estimated
states contain
that the estimated states
statesless well predict
wellharmonic the actual
components,
predict the states.
dueAlso,
actual states. Also,
to high
it is
it is worth
frequency
worth mentioning
switching.
mentioning that that
the the estimated
estimated states
states containless
contain lessharmonic
harmoniccomponents,
components,due due to
to high
high
frequency switching.
frequency switching.

Figure 14. Experimental results for the measured and estimated waveforms of a-phase grid-side
current andExperimental
Figure 14. inverter-side current.
results for the measured and estimated waveforms of a-phase grid-side
Figure 14. Experimental results for the measured and estimated waveforms of a-phase grid-side current
current and inverter-side current.
andevaluate
To inverter-side
the current.
quality of injected currents by the proposed control scheme, Figure 15 shows
the fast Fourier transform
To evaluate the quality(FFT) result currents
of injected for a-phase grid-side
by the current,
proposed controltogether
scheme,with the harmonic
Figure 15 shows
limitsToonevaluate
IEEE the quality
Standard of injected
519-1992. As currents
can be by the
observed, proposed
the THD control
level of scheme,
injected Figure
the fast Fourier transform (FFT) result for a-phase grid-side current, together with the harmoniccurrent15is shows the
relatively
fast Fourier
low. In
limits transform
on fact,
IEEE the (FFT)
injected
Standard result for
currentAshas
519-1992. a-phase
canthe grid-side
THD value
be observed, current,
the of
THD 2.8%,together
levelwhich with the harmonic
successfully
of injected current isfulfillslimits
the
relatively
on IEEE
In Standard
interconnection
low. 519-1992.
standard
fact, the As canaccording
injectedharmonic
current be
hasobserved, the
to IEEE
the THD THD level
Std. of
value 1547. of injected
2.8%, which current is relatively
successfully fulfills low.
the
In fact, the injected
interconnection currentharmonic
standard has the THD value to
according of IEEE
2.8%,Std.
which successfully fulfills the interconnection
1547.
standard harmonic according to IEEE Std. 1547.
Energies 2018, 11, 437 18 of 20
Energies 2018, 11, x FOR PEER REVIEW 17 of 19

Figure 15. Fast Fourier transform (FFT) result for a-phase grid-side current.
Figure 15. Fast Fourier transform (FFT) result for a-phase grid-side current.

6. Conclusions
6. Conclusions
In this paper, a systematic controller design methodology for a three-phase grid-connected
In this
inverter withpaper,
LCLa systematic
filter has beencontroller designbased
presented methodology for a three-phase
on an integral grid-connected
state feedback control and inverter
state
with LCL filter has been presented based on an integral state feedback
observer schemes in a discrete-time domain. To provide a comprehensive design procedure for an control and state observer
schemes in a discrete-time
LCL-filtered grid-connected domain. To provide
inverter, the design a comprehensive
of the proposed design
controlprocedure
scheme for an LCL-filtered
consists of several
grid-connected
steps. First, the accurate system model in the discrete-time domain is developed. Then,steps.
inverter, the design of the proposed control scheme consists of several First,
an integral
the accurate system model in the discrete-time domain is developed. Then,
state feedback controller is designed by using the discrete-time model. In general, to implement the an integral state feedback
controller
integral stateis designed
feedbackbycontroller,
using the thediscrete-time
measurement model.
of allInsystem
general,states
to implement
should bethe integralwhich
required, state
feedback
increases the cost and complexity of system, due to the necessity of additional sensors. In orderthe
controller, the measurement of all system states should be required, which increases to
cost and complexity of system, due to the necessity of additional sensors.
reduce the number of sensors, a discrete-time full-state observer estimates both the inverter-side In order to reduce the
number
currents of andsensors,
capacitora discrete-time
voltages byfull-state
using only observer estimates
the measured both thecurrents
grid-side inverter-side
and gridcurrents and
voltages,
capacitor voltages by using only the measured grid-side currents and grid
which significantly reduces the cost and complexity of the LCL-filtered grid-connected inverter voltages, which significantly
reduces
system in theDG costapplications.
and complexity of the LCL-filtered
In addition, the dynamics grid-connected inverter system
of output currents can be in DG applications.
easily assigned by
In addition, the dynamics of output currents can be easily assigned by
locating the closed-loop poles to desired values. Due to the discrete-time integrator incorporatedlocating the closed-loop poles
in
to desired values. Due to the discrete-time integrator incorporated in
the state feedback controller, the proposed control scheme ensures both the reference tracking and the state feedback controller,
the proposedrejection
disturbance control scheme ensuresin
performance both the reference
a practical tracking
and simple and Aside
way. disturbance
from therejection performance
detailed analyses,
in a practical and simple way. Aside from the detailed analyses,
the proposed scheme provides a systematic and straightforward controller design procedure. the proposed scheme provides a
systematic
To verifyand straightforward controller design
the practical usefulness of theprocedure.
proposed control scheme, a 2 kW prototype
To verify the practical usefulness
grid-connected inverter has been constructed, and of the proposed controlthescheme,
proposed a 2 kW prototype
control grid-connected
system has been
inverter has been constructed, and the proposed control
implemented on 32-bit floating-point DSP TMS320F28335. Comprehensive simulation system has been implemented on 32-bit
and
floating-point DSP TMS320F28335. Comprehensive simulation and experimental
experimental results are presented, to prove the feasibility and effectiveness of the proposed scheme. results are presented,
to
Theprove the feasibility
transient response of andtheeffectiveness
proposed scheme of theisproposed scheme.theThe
very fast, having transient
settling time response
of about 0.5 of ms.
the
proposed scheme is very fast, having the settling time of about 0.5
The injected currents by the proposed control scheme successfully fulfill the interconnection ms. The injected currents by the
proposed
standard control
harmonic scheme successfully
specified by IEEEfulfill
Std. the
1547,interconnection
and have a low standard
THD valueharmonic specified
of 2.8%. by IEEE
By virtue of
Std. 1547, and have a low THD value of 2.8%. By virtue of simple and
simple and low-cost implementation, as well as systematic control design, the proposed control low-cost implementation, as
well
schemeas systematic control way
can be a practical design, the proposed
of designing control scheme
a grid-connected can bein
inverter a DG
practical way of designing
applications. In addition a
grid-connected
to providing a basic framework to synthesize a controller for a grid-connected inverter in aa
inverter in DG applications. In addition to providing a basic framework to synthesize
controller
discrete-timefor astate
grid-connected inverter inwork
space, the proposed a discrete-time
can contributestate to
space, the proposed
promote succeeding work can contribute
research studies,
to promote succeeding research studies, such as the optimal control design
such as the optimal control design or grid frequency-adaptive control, because they are often or grid frequency-adaptive
control,
designed because they are often
in a discrete-time designed
state model. in a discrete-time state model.

Acknowledgments: This research was supported by Basic Science Research Program through the National
Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2016R1D1A1B03930975). This
work was also supported by the Human Resources Development of the Korea Institute of Energy Technology
Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry &
Energies 2018, 11, 437 19 of 20

Acknowledgments: This research was supported by Basic Science Research Program through the National
Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2016R1D1A1B03930975). This
work was also supported by the Human Resources Development of the Korea Institute of Energy Technology
Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry & Energy
(NO. 20154030200720).
Author Contributions: Seung-Jin Yoon, Ngoc Bao Lai and Kyeong-Hwa Kim conceived the main concept of the
control structure and developed the entire system. Seung Jin Yoon and Ngoc Bao Lai carried out the research
and analyzed the numerical data with the guidance from Kyeong-Hwa Kim. Seung Jin Yoon, Ngoc Bao Lai and
Kyeong-Hwa Kim collaborated to prepare the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. De Matos, J.; Silva, F.; Ribeiro, L. Power control in AC isolated microgrids with renewable energy sources
and energy storage systems. IEEE Trans. Ind. Electron. 2014, 62, 3490–3497. [CrossRef]
2. Han, H.; Hou, X.; Yang, J.; Wu, J.; Su, M.; Guerrero, J.M. Review of power sharing control strategies for
islanding operation of ac microgrids. IEEE Trans. Smart Grid 2015, 7, 200–215. [CrossRef]
3. Lasseter, R.; Akhil, A.; Marnay, C.; Stephens, J.; Dagle, J.; Guttromson, R.; Meliopoulous, A.S.; Yinger, R.;
Eto, J. The CERTS Microgrid Concept, White Paper on Integration of Distributed Energy Resources; U.S. Department
of Energy: Washington, DC, USA, 2002.
4. Teodorescu, R.; Liserre, M.; Rodriguez, P. Grid Converters for Photovoltaic and Wind Power Systems; Wiley:
Hoboken, NJ, USA, 2011; pp. 314–320.
5. IEEE Standards Association. IEEE Std. 1547-2003. Standard for Interconnecting Distributed Resources with
Electric Power Systems; IEEE Standards Association: Piscataway, NJ, USA, 2003.
6. Bouzid, A.M.; Guerrero, J.M.; Cheriti, A.; Bouhamida, M.; Sicard, P.; Benghanem, M. A survey on control of
electric power distributed generation systems for microgrid applications. Renew. Sustain. Energy Rev. 2015,
44, 751–766. [CrossRef]
7. Nguyen, T.H.; Kim, K.H. Finite control set—Model predictive control with modulation to mitigate harmonic
component in output current for a grid-connected inverter under distorted grid conditions. Energies 2017, 10,
907. [CrossRef]
8. Trinh, Q.N.; Lee, H.H. An advanced current control strategy for three-phase shunt active power filters. IEEE
Trans. Ind. Electron. 2013, 60, 5400–5410. [CrossRef]
9. Liu, F.; Zha, X.; Zhou, Y. Design and research on parameter of LCL filter in three-phase grid-connected
inverter. In Proceedings of the IEEE International Conference on Power Electronics and Motion Control,
Wuhan, China, 19–20 May 2009.
10. Blaabjerg, F.; Teodorescu, R.; Liserre, M.; Timbus, A.V. Overview of control and grid synchronization for
distributed power generation systems. IEEE Trans. Ind. Electron. 2006, 53, 1398–1409. [CrossRef]
11. Heo, H.S.; Choe, G.H.; Mok, H.S. Robust predictive current control of a grid-connected inverter with
harmonics compensation. In Proceedings of the IEEE International Conference on Power Electronics
Conference and Exposition, Long Beach, LA, USA, 17–21 March 2013.
12. Abusara, M.A.; Jamil, M.; Sharkh, S.M. Repetitive current control of an interleaved grid-tied inverter.
In Proceedings of the IEEE International Symposium on Power Electronics for Distributed Generation
Systems (PEDG), Aalborg, Denmark, 25–28 June 2012.
13. Chen, D.; Zhang, J.; Zhang, Z. An improved repetitive control scheme for grid-connected inverter with
frequency-adaptive capability. IEEE Trans. Ind. Electron. 2013, 60, 814–823. [CrossRef]
14. Liu, T.; Hao, X.; Yang, X.; Zhao, M.; Huang, Q.G.; Huang, L. A novel repetitive control scheme for three-phase
grid-connected inverter with LCL filter. In Proceedings of the 7th International Power Electronics and Motion
Control Conference, Harbin, China, 2–5 June 2012.
15. Teodorescu, R.; Blaabjerg, F.; Liserre, M.; Loh, P.C. Proportional-resonant controllers and filters for
grid-connected voltage-source converters. IEE Proc. Electr. Power Appl. 2006, 153, 750–762. [CrossRef]
16. Abusara, M.A.; Sharkh, S.M. Design of a robust digital current controller for a grid connected interleaved
inverter. In Proceedings of the IEEE International Symposium on Industrial Electronics, Bari, Italy,
4–7 July 2010.
Energies 2018, 11, 437 20 of 20

17. Han, Y.; Shen, P.; Zhao, X.; Guerrero, J. Control strategies for islanded microgrid using enhanced hierarchical
control structure with multiple current-loop damping schemes. IEEE Trans. Smart Grid 2017, 8, 1139–1153.
[CrossRef]
18. Gonzatti, R.B.; Ferreira, S.C.; da Silva, C.H.; Pereira, R.R.; da Silva, L.E.B.; Lambert-Torres, G. Using smart
impedance to transform high impedance microgrid in a quasi-infinite busbar. IEEE Trans. Smart Grid 2017, 8,
428–436. [CrossRef]
19. Perez-Estevez, D.; Doval-Gandoy, J.; Yepes, A.G.; Lopez, O.; Baneira, F. Enhanced resonant current controller
for grid-connected converters with LCL filter. IEEE Trans. Power Electron. 2017. [CrossRef]
20. Liu, Y.; Wu, W.; He, Y.; Lin, Z.; Blaabjerg, F.; Chung, H.S.H. An efficient and robust hybrid damper for LCL
or LLCL-based grid-tied inverter with strong grid-side harmonic voltage effect rejection. IEEE Trans. Ind.
Electron. 2016, 63, 926–936. [CrossRef]
21. Jin, W.; Li, Y.; Sun, G.; Bu, L. H∞ repetitive control based on active damping with reduced computation
delay for LCL-type grid-connected inverters. Energies 2017, 10, 586. [CrossRef]
22. Fu, X.; Li, S. A novel neural network vector control for single-phase grid-connected converters with L,
LC and LCL filters. Energies 2017, 9, 328. [CrossRef]
23. TMS320F28335 Digital Signal Controller (DSC)—Data Manual; Texas Instrument: Dallas, TX, USA, 2008.

© 2018 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/).