[Link].

org
Published in IET Renewable Power Generation
Received on 12th September 2011
Revised on 29th March 2012
Accepted on 6th July 2012
doi: 10.1049/iet-rpg.2011.0243

ISSN 1752-1416

Simple synchronisation technique for three-phase

grid-connected distributed generation systems
Xiao-Qiang Guo, Wei-Yang Wu
Key Laboratory of Power Electronics for Energy Conservation and Motor Drive of Hebei Province, Department of Electrical
Engineering, Yanshan University, Qinhuangdao, People’s Republic of China
E-mail: gxq@[Link]

Abstract: Grid synchronisation is crucial for integration of renewable energy sources and distributed generation systems into the
grid. Accurate and fast estimation of the fundamental positive sequence component is a necessity for synchronisation of three-
phase distributed generation systems with the grid. In this study, a simple synchronisation technique based on the real-
complex-coefficient filters is proposed. It has a very simple structure, and can directly estimate the fundamental positive and
negative sequences under the distorted, unbalanced and offset conditions with no need of a phase-locked loop. Furthermore,
the dc offset can also be extracted precisely. For optimising the steady-state and dynamic performance, a parameter tuning
guidance is provided from the eigenvalue analysis point of view. In addition, a simple auxiliary unit is presented to remain its
frequency-adaptive feature. Finally, experimental results based on a 32-bit fixed-point TMS320F2812 DSP verify the
effectiveness of the proposed synchronisation method.

1 Introduction Therefore the fundamental and its quadrature components

can be obtained, and the FPS can be easily obtained with
Integration of renewable energy sources and distributed the symmetrical component method in the stationary αβ
generation into the microgrid or smart grid is getting more frame. Aside from DSOGI, two interesting non-linear
and more attention because of the environmental concerns solutions have been reported in [14–17], named enhanced
and electric utility deregulation [1–8]. However, there are phase-locked loop and adaptive notch filter (ANF), which
many technical challenges that should be settled for the can output each phase fundamental voltage and its
successful grid integration of renewable energy sources. quadrature component. With the information, the FPS can
One of the most important issues is the synchronisation be calculated with the instantaneous symmetrical
technique, where an accurate and fast estimation of the component method. Synchronous rotating frame PLL (SRF
fundamental positive sequence (FPS) is a necessity, PLL) is also a commonly used approach to FPS estimation.
especially for the fault ride through requirement [9, 10]. It has the fast and accurate performance for sinusoidal and
Many estimation algorithms for FPS have been developed balanced input voltages. Under the harmonics and
in the past decade. The simplest one is the delayed signal unbalanced conditions, however, its bandwidth has to be
cancellation method, which can achieve the accurate reduced to a certain level for the accurate estimation of FPS
estimation of FPS with a time delay of one-quarter of a at the cost of transient response [18], which is not suitable
period. However, some actions should be taken to ensure for real-time control of three-phase grid-connected DG. In
the accuracy when the grid frequency fluctuates [11], and it despite of the limitation, SRF PLL is still widely used
is sensitive to harmonics. An alternative solution is the because of its simplicity, and sometimes used as an
symmetrical component method [12], which is a basic auxiliary unit to provide the information such as the
analysis tool for power system faults. However, it is amplitude, phase and frequency. An enhanced version of
sensitive to harmonics, and it requires phase-shift SRF PLL is the double-synchronous rotating frame PLL
information, which complicates the implementation. Several [19, 20], which uses a unique decoupling network to extract
modified methods have been presented for improving the the FPS with no need of the bandwidth reduction. However,
performance of the symmetrical-component-based method. the computational burden is increased because of many
The basic idea behind them is to filter the harmonics and rotating frame transformations.
output the fundamental and its 90° shift components. The In this paper, a real-complex-coefficient filter structure is
dual second-order generalised integrator (DSOGI) [13] presented. In contrast with the aforementioned methods, the
proposed by Rodríguez is a typical one, which is a smart proposed method can precisely estimate the FPS in a
combination of a low-pass filter (LPF) and a band-pass frequency-adaptive way with no need of signal delay,
filter (BPF). BPF provides harmonic-filtering functions, and symmetrical component method or rotating frame
LPF offers both the harmonic filtering and 90° phase shift. transformation. Furthermore, it has an interesting feature of

IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62 55

doi: 10.1049/iet-rpg.2011.0243 & The Institution of Engineering and Technology 2013
 17521424, 2013, 1, Downloaded from [Link] by Algeria Hinari NPL, Wiley Online Library on [16/02/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link]
offset rejection, which cannot be settled for the majority of the
grid synchronisation methods [21].

2 Proposed synchronisation technique

This section will present the basic principle and practical
implementation of the proposed PLL-less synchronisation
method.

2.1 Mathematic model

Considering a three-phase three-wire system, where zero

sequence and harmonic components are omitted, the
three-phase voltage can be generally expressed as follows
(see (1))

where superscripts + and – denote the positive and negative

sequences.U1+ , ω1 and w+
1 are the amplitude, frequency and
phase angle of the positive-sequence voltages, respectively.
U1− and w− 1 are the amplitude and phase angle of the Fig. 1 RCCF structure
negative-sequence voltages, respectively. Ua0, Ub0, Uc0 are
dc components.
Using the Clarke transformation, (1) can be expressed as
The frequency-domain expression of the proposed RCCF
⎡ ⎤ structure can be obtained from Fig. 1 as follows
Ua  +
Ua Ua−




Ua U a0
= T ⎣ Ub ⎦ = + − + (2)
Ub Ub+ Ub U b0 ⎧
v0
Uc ⎪
⎪ Ûa0 (s) = Ua (s) − Ûa+ (s) − Ûa− (s)

⎪
⎪
⎪ s + v0
v0 
⎪
⎪ 
where Ûb0 (s) = Ub (s) − Ûb+ (s) − Ûb− (s)
⎪
⎪
⎪
⎪
⎪
⎪ s + v0
vc
⎪
⎪ + −
⎪ 
⎨ Ûa (s) = s − jv + v Ua (s) − Ûa0 (s) − Ûa (s)
⎡ ⎤ ⎪
1 1 ⎪
1 − − ⎥ 1 c
(3)
2⎢
√2 √2 ⎥ vc 
T= ⎢ + −
⎪ Ûb (s) = U (s) − Û b0 (s) − Û (s)
3⎣ 3 3⎦ s − jv1 + vc b b
⎪
⎪
⎪
0 − ⎪
⎪
v
2 2
⎪
Ûa− (s) = c
Ua (s) − Ûa0 (s) − Ûa+ (s)
⎪
⎪ 
⎪
s + j v1 + vc
⎪
⎪
⎪
⎪
⎪
− vc  
Ub (s) − Ûb0 (s) − Ûb+ (s)
 
Ua+
⎪
U1+ sin(v1 t + w+
1) ⎩ Ûb (s) =
⎪
⎪
= s + jv1 + vc
Ub+ −U1+ cos(v1 t + w+
1)

Ua− U1− sin(− v1 t + w−


1)
= The corresponding time-domain expression for (3) is
Ub− −
−U1 cos(− v1 t + w− 1)

U a0


1 2Ua0 − Ub0 − Uc0 ⎧  
Û˙ a0 (t) = v0 Ua (t) − Û a0 (t) − Û a (t) − Û a (t)
+ −
= √ ⎪
U b0 3 3(Ub0 − Uc0 ) ⎪
⎪
⎪
⎪  
Û˙ b0 (t) = v0 Ub (t) − Û b0 (t) − Û b (t) − Û b (t)
⎪
⎪ + −
⎪
⎪
⎪
Ua+ , Ub+ , Ua− , Ub− , Uα0 Uβ0 are the positive sequence,
⎪
⎪ +  
Û˙ a (t) = vc Ua (t) − Û a0 (t) − Û a (t) − Û a (t) − v1 Û b (t)
+ − +
⎪
⎪
⎪
negative sequence and dc components of Uα and Uβ in ⎨
stationary αβ frame, respectively. +  
Û˙ b (t) = vc Ub (t) − Û b0 (t) − Û b (t) − Û b (t) + v1 Û a (t)
⎪ + − +
To extract the FPS components from the grid voltage
⎪
⎪
⎪
⎪
with immunity to the other components, a
⎪ −  
Û˙ a (t) = vc Ua (t) − Û a0 (t) − Û a (t) − Û a (t) + v1 Û b (t)
⎪
⎪ + − −
⎪
real-complex-coefficient-filter (RCCF) structure is proposed ⎪
⎪
⎪
+ − ⎪ −
as shown in Fig. 1, where Ûab0 , Ûab and Ûab are the
 
Û˙ b (t) = vc Ub (t) − Û b0 (t) − Û b (t) − Û b (t) − v1 Û a (t)
⎪
⎪
⎩ + − −
estimated dc offset, fundamental positive- and
negative-sequence components. (4)

Ua0 + U1+ sin(v1 t + w+ − −

⎡ ⎤ ⎡ ⎤
Ua 1 ) + U1 sin(− v1 t + w1 )
⎣ Ub ⎦ = ⎣ U + U + sin(v t + w+ − 120W ) + U − sin(− v t + w− − 120W ) ⎦ (1)
b0 1 1 1 1 1 1
Uc Uc0 + U1+ sin(v1 t + w+ W − − W
1 + 120 ) + U1 sin(− v1 t + w1 + 120 )

56 IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62

& The Institution of Engineering and Technology 2013 doi: 10.1049/iet-rpg.2011.0243
 17521424, 2013, 1, Downloaded from [Link] by Algeria Hinari NPL, Wiley Online Library on [16/02/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link]
The state-space model of the RCCF structure can be derived negative real parts, and the oscillatory frequency in transient
from (4) as (5). response depends on the imaginary parts of dominant
complex-conjugate eigenvalues. Besides, the further the

ẋ(t) = A(t) · x(t) + B(t) · u(t) left-half-plane eigenvalues are away from the imaginary
(5) axis, the faster the transient response is.
y(t) = C · x(t)
To simplify the parameter tuning procedure, (5) can be
 T written in a reduced-order form from the viewpoint of
where x(t) = y(t) = Ûa0 Ûb0 Ûa+ Ûb+ Ûa− Ûb− , vector (i.e. Uαβ = Uα+ jUβ), instead of scalar. Where (t) =
T
u(t) = Ua Ub Ua Ub Ua Ub + −
T
y(t) = Ûab0 Ûab Ûab
−v0 0 −v 0 0 −v 0 0
⎡ ⎤
 T
+ −
⎢ 0 −v 0 0 −v0 0 −v 0 ⎥ x(t) = y(t) = Ûab0 Ûab Ûab
⎢ ⎥
⎢ ⎥
⎢ −v c 0 −v c −v1 −v c 0 ⎥ u(t) = Uab Uab Uab
T
A(t) = ⎢
⎢ 0
⎥
⎢ −vc v 1 −v c 0 −v c ⎥
⎥
⎢ ⎥
⎣ −v c 0 −v c 0 −v c v1 ⎦
⎡ ⎤
0 −vc 0 −v c −v 1 −v c 6×6
−v 0 −v 0 −v 0
A(t) = ⎣ −vc −vc + jv1 −vc

 ⎢ ⎥
v0 I2×2 0 ⎦
B(t) = , C = I6×6 −vc −v c −vc − jv1
0 vc I4×4 3×3
⎡ ⎤
v0 0 0
The time-domain solution of (5) can be written as
B(t) = ⎣ 0 vc 0 ⎦ , C = I3×3
⎢ ⎥
t 0 0 vc 3×3
x(t) = eA(t−t0 ) x(t0 ) + eA(t−t) Bu(t) dt (6)
t0
Fig. 2 shows the system eigenvalues diagram with the
The steady-state solution of (6) can be solved as variation of ωc under the condition that ω0 = 0. It can be
observed that the eigenvalues tend to be far away from the
⎡ ⎤ ⎡ imaginary axis and close to the real axis by increasing ωc
Ûa0 U a0
⎤
from 10 to 314 (turning point), which indicates that the
⎢ Ûb0 ⎥ ⎢ Ub0 ⎥
⎢ ⎥ stability and transient response are enhanced. However,
⎢ +⎥ ⎢ ⎥
⎢ Ûa ⎥ ⎢
x(t) = ⎢ + ⎥ = ⎢ Ua+ ⎥ ⎥ when ωc is increased after the turning point, the eigenvalues
⎢ Ûb ⎥ ⎢ Ub+ ⎥ are spilt, and one tends to be close to the origin, becoming
⎢ −⎥ ⎣ −⎥ the dominant, which reveals that the stability and transient
⎢
⎣ Û ⎦ Ua ⎦
a response are deteriorated. Considering the tradeoff between
Ûb− Ub−
the system transient response and filtering performance, the
parameter of ωc is chosen as 314.
⎡ ⎤
(2U√a0 − Ub0 − Uc0 )/3
⎢ 3(Ub0 − Uc0 )/3 ⎥ On the other hand, assuming that ωc is fixed to 314, the
⎢ U1+ sin(v1 t + w+
⎢ ⎥ system eigenvalues diagram with the variation of ω0 is
1) ⎥
⎥
=⎢⎢ + + ⎥ (7) depicted in Fig. 3. It can be observed that the dominant
−U
⎢ −1 cos( v 1 t + w 1 )
⎣ U1 sin(− v1 t + w−
⎥ eigenvalues tend to be far away from the imaginary axis by
1)
⎦
increasing ω0 from 10 to the turning point ( ∼ 42.35).
−U1− cos(− v1 t + w− 1 ) However, increasing ω0 after the turning point will make
the dominant eigenvalues move to the imaginary axis and
From (7), it is clear that the proposed RCCF structure can far away from the real axis, which indicates the stability
accurately extract the FPS component in presence of the and transient response are deteriorated. So the value of ω0
negative sequence and dc components, which can be also
estimated precisely. Note that, in practical applications, the
grid voltage may suffer from the harmonic components.
Considering the filtering characteristics of the RCCF
structure, the effect of harmonics can be alleviated by
tuning the filter parameters with low cutoff frequency.
However, it would degrade the dynamic performances of
the RCCF structure. Therefore a balance should be taken
between the steady-state and dynamic performances.

2.2 Parameter tuning

As mentioned before, the parameters of the RCCF structure

should be carefully designed to keep the harmonics better
attenuated. On the other hand, the stability and dynamic
response should be also ensured. It is well known that the
eigenvalue analysis is an effective tool to evaluate the
system stability and transient response. Generally, the
system stability can be confirmed if all the eigenvalues have Fig. 2 Eigenvalues diagram with ω0 = 0

IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62 57

doi: 10.1049/iet-rpg.2011.0243 & The Institution of Engineering and Technology 2013
 17521424, 2013, 1, Downloaded from [Link] by Algeria Hinari NPL, Wiley Online Library on [16/02/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link]

Fig. 3 Eigenvalues diagram with ωc = 314

is selected as 42.35 for a good tradeoff between the filtering

performance and transient response. Note that, the dominant
eigenvalues will be closer to the imaginary axis when ω0 is Fig. 4 Implementation of RCCF structure
changed from 0 (Fig. 2) to 42.35 (Fig. 3) for the effective
dc offset rejection or measurement, but it will lead to a
slow transient, which will be verified in the following say, the operating orbit will move towards the steady-state
sections. operating point of ω1, and vice versa. Therefore the stability
of the frequency estimation process will be guaranteed.
2.3 Structure implementation
3 Performance evaluation
Fig. 4 shows the detailed implementation of the proposed
RCCF structure, which consists of several very simple To evaluate the performance of the proposed method, the
mathematical calculations such as sum, multiplication and time-domain simulations are carried out in the Matlab/
integration. Therefore it can be easily implemented. Note Simulink environment. Note that Yazdani et al. [22] have
that a simple auxiliary unit (dash line) is added to the provided a performance comparison with different
RCCF structure in order to ensure the accuracy under the synchronisation techniques, which indicates that ANF is
variable-frequency conditions. That is, the frequency ω1 in better choice than others. Therefore we perform
Fig. 4 should be updated with v̂1 to remain the comparisons with ANF in the paper.
frequency-adaptive feature.
The mathematic equation of the frequency estimation in
Fig. 4 can be expressed as follows 3.1 Unbalance and harmonics

 +
Initial parameters are set as follows: three-phase grid voltage
k Ûb+ v̂1 Ûb+ + Û˙ a is balanced and its line–line RMS amplitude and fundamental
v̂˙ 1 = − (8) frequency are 1.0 pu and 50 Hz, respectively. The parameters
vc of the proposed RCCF structure are set as ωc = 314 and k = 40
000. Considering that there is no dc component, the parameter
of ω0 is set to zero for simplicity. Then, the negative
Considering the input as (1), the steady-state solution of (8) sequences, fifth harmonic and seventh harmonic
can be solved as (9). components are superposed on the initial voltage with the
⎡ ⎤ ⎡ ⎤ values of 0.4, 0.04 and 0.03 pu, respectively, where the
v̂1 v1 total harmonic distortion of the input voltage is 5%.
⎣ Ûa+ ⎦ = ⎣ U1+ sin(v1 t + w+
1)
⎦ (9) Besides, the grid frequency is changed from 50 to 55 Hz to
+ + +
Ûb −U1 cos(v1 t + w1 ) evaluate the extraction accuracy under the frequency
excursion condition.
Fig. 5 shows the simulation results, from which it can be
+
Around the steady-state operating point of (9), Û˙ a ≃ − v1 Ûb+ , observed that the proposed method can accurately extract
so (8) can be rewritten as the positive, negative sequences within one cycle (20 ms).
Besides, the frequency of the positive sequence can also be
estimated precisely. On the other hand, the simulation
k(v̂1 − v1 )(Ûb+ )2 results with three-phase ANF are also provided for purpose
v̂˙ 1 = − (10)
vc of performance comparison. The parameters of ANF are
tuned according to [16]. As shown in Fig. 6, three-phase
From (10), it can be observed that when the estimation ANF can also achieve the fast and accurate estimation of
frequency of v̂1 deviates from the steady-state operating sequence information, and the transient response time is
point of ω1 to a high value (e.g. v̂1 − v1 . 0), its rate of comparable with the proposed RCCF structure. Note that
change will be a negative value (e.g. v̂˙ 1 , 0). That is to the proposed method can achieve the direct estimation with

58 IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62

& The Institution of Engineering and Technology 2013 doi: 10.1049/iet-rpg.2011.0243
 17521424, 2013, 1, Downloaded from [Link] by Algeria Hinari NPL, Wiley Online Library on [16/02/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link]

Fig. 5 Simulation results with proposed method under unbalanced

and harmonic conditions

Fig. 7 Simulation results with proposed method under

unbalanced, harmonic and dc offset conditions

no need of instantaneous symmetrical component method,

which is a necessity in three-phase ANF.

3.2 Unbalance, harmonics and dc offset

The following will evaluate the performance under the

unbalance, harmonics and dc offset conditions. The
parameters remain unchanged with the previous case except
that the dc offset (Ua0 = −0.03, Ub0 = 0.05, Uc0 = −0.02) is
superposed on the initial voltage, and ω0 is set to 42.35.
From Fig. 7, it is clear that the proposed RCCF
structure can extract the fundamental positive/negative
sequence components and its frequency in a precise and
frequency-adaptive way. Also, the dc offset can be
extracted accurately in presence of high levels of
fundamental positive/negative sequences and harmonic
components, thanks to the unique RCCF structure and its
filtering feature.
On the other hand, three-phase ANF is sensitive to dc
offset, and the estimated frequency suffers from the
low-frequency ripples, as shown in Fig. 8. It should be
noted that the transient response (about two cycles) of the
Fig. 6 Simulation results with ANF under unbalanced and proposed method is not as fast as three-phase ANF in this
harmonic conditions case. The reason is that the dominant eigenvalues will be

IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62 59

doi: 10.1049/iet-rpg.2011.0243 & The Institution of Engineering and Technology 2013
 17521424, 2013, 1, Downloaded from [Link] by Algeria Hinari NPL, Wiley Online Library on [16/02/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link]

Fig. 8 Simulation results with ANF under unbalanced, harmonic

and dc offset conditions

closer to the imaginary axis when ω0 is changed from 0 to

42.35 for the effective offset rejection, which is in
agreement with the theoretical analysis in Section 2 (see
Figs. 2 and 3). Therefore in practice, it is recommended
that the parameter of ω0 is set to zero for the fast transient
response, on condition that the input voltage is free of dc
offset. On the other hand, the parameter of ω0 should be
set for the accurate estimation of the input voltage
information in presence of dc offset at the expense of
transient response.
It should be noted that the comparison is carried out only
between the proposed method and three-phase ANF. A
comprehensive performance comparison with other grid
synchronisation algorithms will be reported in a future paper.

4 Experimental results Fig. 9 Experimental results under unbalanced and harmonic

conditions
This section will provide the discrete-domain implementation a Input voltages
of the proposed RCCF structure. The experimental tests are b Estimated fundamental positive/negative sequences
carried out based on a 32-bit fixed-point 150 MHz c Estimated phase angle and frequency
TMS320F2812 DSP with a sample rate of 10 kHz. The
GLOBAL_Q is defined as 22 (IQ22). The parameters of the
this paper.
proposed RCCF structure are set as ωc = 314 and k = 40
000. An arbitrary three phase waveform generator based on
a DSP is used to generate the simulated input voltage. y(s) 1 y(z) Ts 23z−1 − 16z−2 + 5z−3
There are many discretisation methods such as forward/ = ⇔ = (11)
u(s) s u(z) 12 1 − z−1
backward Euler method, trapezoidal method, second-order
integrator and third-order integrator. A discretisation
Ts
accuracy comparison of abovementioned methods has been y(n) = y(n − 1) + [23u(n − 1) − 16u(n − 2) + 5u(n − 3)]
reported in [23], which indicates that the third-order 12
integrator (TOI) is the best one. Therefore TOI is used in (12)

60 IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62

& The Institution of Engineering and Technology 2013 doi: 10.1049/iet-rpg.2011.0243
 17521424, 2013, 1, Downloaded from [Link] by Algeria Hinari NPL, Wiley Online Library on [16/02/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link]
where Ts is the sample period, u and y are the integrator input
and output, respectively.
The experimental results are shown in Figs. 9 and 10. It can
be observed that there is a good agreement between
simulation results and experimental ones.

5 Conclusion
This paper has presented a simple grid synchronisation
method based on the real-complex-coefficient filters.
Different from the existing ideas, with no need of the signal
delay, symmetrical component method or rotating frame
transformations, the fundamental positive/negative sequences
and dc components can be accurately estimated in a
frequency-adaptive way under the distorted, unbalanced and
offset conditions. In addition, it has a very simple structure,
which can be easily implemented by analog devices or
digital signal processors. These interesting features make it
attractive for flexible operation of three-phase grid-connected
distributed generation system under non-ideal conditions.

6 Acknowledgments
This work was supported by the National Natural Science
Foundation of China (grant numbers 50837003, 50977081)
and Hebei Province Natural Science Foundation (grant
number E2012203023).

7 References
1 Li, H., Chen, Z.: ‘Overview of different wind generator systems and
their comparisons’, IET Renew. Power Gener., 2008, 2, (2), pp. 123–138
2 Mathiesen, B.V., Lund, H.: ‘Comparative analyses of seven
technologies to facilitate the integration of fluctuating renewable
energy sources’, IET Renew. Power Gener., 2009, 3, (2), pp. 190–204
3 Geng, H., Yang, G.: ‘Robust pitch controller for output power levelling
of variable-speed variable-pitch wind turbine generator systems’, IET
Renew. Power Gener., 2009, 3, (2), pp. 168–179
4 Madureira, A.G., Peças Lopes, J.A.: ‘Coordinated voltage support in
distribution networks with distributed generation and microgrids’, IET
Renew. Power Gener., 2009, 3, (4), pp. 439–454
5 Majumder, R., Ghosh, A., Ledwich, G., Zare, F.: ‘Load sharing and
power quality enhanced operation of a distributed microgrid’, IET
Renew. Power Gener., 2009, 3, (2), pp. 109–119
6 Hajizadeh, A., Golkar, M.A.: ‘Fuzzy neural control of a hybrid fuel cell/
battery distributed power generation system’, IET Renew. Power Gener.,
2009, 3, (4), pp. 402–414
7 Saha, A.K., Chowdhury, S., Chowdhury, S.P., Song, Y.H.: ‘Application
of solid-oxide fuel cell in distributed power generation’, IET Renew.
Power Gener., 2007, 1, (4), pp. 193–202
8 Guo, X.Q., Wu, W.Y.: ‘Improved current regulation of three-phase
grid-connected voltage source inverters for distributed generation
systems’, IET Renew. Power Gener., 2010, 4, (2), pp. 101–115
9 Busquets-Monge, A.S., Bordonau, J., Martinez Velasco, A., Silva, C.,
Pontt, J., Rodriguez, J.: ‘Control strategies based on symmetrical
components for grid-connected converters under voltage dips’, IEEE
Trans. Ind. Electron., 2009, 56, (6), pp. 2162–2173
10 Benz, C.H., Franke, W.T., Fuchs, F.W.: ‘Low voltage ride through
capability of a 5 kW grid-tied solar inverter’. Proc. EPE/PEMC, 2010,
pp. 13–20
11 Svensson, J., Bongiorno, M., Sannino, A.: ‘Practical implementation of
delayed signal cancellation method for phase-sequence separation’,
IEEE Trans. Power Deliv., 2007, 22, (1), pp. 18–26
12 Paap, G.C.: ‘Symmetrical components in the time domain and their
application to power network calculations’, IEEE Trans. Power Syst.,
2000, 15, (2), pp. 522–528
Fig. 10 Experimental results under unbalanced, harmonic and dc 13 Rodriguez, P., Teodorescu, R., Candela, I., Timbus, A.V., Liserre, M.,
offset conditions Blaabjerg, F.: ‘New positive-sequence voltage detector for grid
synchronization of power converters under faulty grid conditions’.
a Input voltages Proc. IEEE PESC, 2006, pp. 1–7
b Estimated fundamental positive and negative sequences 14 Ghartemani, M.K., Iravani, M.R.: ‘A method for synchronization of
c Estimated dc offset power electronic converters in polluted and variable-frequency
d Estimated phase angle and frequency environments’, IEEE Trans. Power Syst., 2004, 19, (3), pp. 1263–1270

IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62 61

doi: 10.1049/iet-rpg.2011.0243 & The Institution of Engineering and Technology 2013
 17521424, 2013, 1, Downloaded from [Link] by Algeria Hinari NPL, Wiley Online Library on [16/02/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link]
15 Yazdani, D., Bakhshai, A., Joos, G., Mojiri, M.: ‘A nonlinear adaptive converters control’, IEEE Trans. Power Electron., 2007, 22, (2),
synchronization technique for grid-connected distributed energy’, pp. 584–592
IEEE Trans. Power Electron., 2008, 23, (4), pp. 2181–2186 20 Rodriguez, P., Timbus, A., Teodorescu, R., Liserre, M., Blaabjerg, F.:
16 Yazdani, D., Mojiri, M., Bakhshai, A., Joos, G.: ‘A fast and accurate ‘Reactive power control for improving wind turbine system behavior
synchronization technique for extraction of symmetrical components’, under grid faults’, IEEE Trans. Power Electron., 2009, 24, (7),
IEEE Trans. Power Electron., 2009, 24, (3), pp. 674–684 pp. 1798–1801
17 Yazdani, D., Bakhshai, A., Joos, G., Mojiri, M.: ‘A real-time 21 Ciobotaru, M., Teodorescu, R., Agelidis, V.G.: ‘Offset rejection for PLL
extraction of harmonic and reactive current in a nonlinear load for based synchronization in grid-connected converters’. Proc. IEEE APEC,
grid-connected converters’, IEEE Trans. Ind. Electron., 2009, 56, (6), 2008, pp. 1611–1617
pp. 2185–2189 22 Yazdani, D., et al.: ‘Three-phase grid synchronization techniques for
18 Chung, S.-K.: ‘A phase tracking system for three phase utility grid connected converters in distributed generation systems’. Proc.
interface inverters’, IEEE Trans. Power Electron., 2000, 15, (3), IEEE ISIE, 2009, pp. 1105–1110
pp. 431–438 23 Ciobotaru, M., Teodorescu, R., Blaabjerg, F.: ‘A new single phase PLL
19 Rodriguez, P., Pou, J., Bergas, J., Candela, J., Burgos, R., Boroyevich, structure based on second order generalized integrator’. Proc. IEEE
D.: ‘Decoupled double synchronous reference frame PLL for power PESC, 2006, pp. 1–6

62 IET Renew. Power Gener., 2013, Vol. 7, Iss. 1, pp. 55–62

& The Institution of Engineering and Technology 2013 doi: 10.1049/iet-rpg.2011.0243