10 1016@j Solener 2020 06 085

Solar Energy 207 (2020) 851–873
Contents lists available at ScienceDirect
Solar Energy
journal homepage: www.elsevier.com/locate/solener
A review on current injection techniques for low-voltage ride-through and T

grid fault conditions in grid-connected photovoltaic system
Z. Hassana, , A. Amira, J. Selvaraja, , N.A. Rahima,b
⁎ ⁎
a
Higher Institution Centre of Excellence (HICoE), UM Power Energy Dedicated Advanced Centre (UMPEDAC), Level 4, Wisma R&D, University of Malaya, Jalan Pantai
Baharu, 59990 Kuala Lumpur, Malaysia
b
Renewable Energy Research Group, King Abdulaziz University, Jeddah 21589, Saudi Arabia
ARTICLE INFO ABSTRACT
Keywords: In order to meet the increasing energy demands of the modern-day world, the shift towards renewable energy
Grid-connected photovoltaic (GCPV) has proved to be a viable alternative, as it has significantly reduced reliance on conventional energy resources.
Low-voltage ride-through (LVRT) Solar energy, in particular, remains one of the best available renewable energy options, as it is abundant, clean
Grid codes (GCs) and reliable. Owing to the susceptibility of grid-connected photovoltaic (GCPV) system against grid faults,
Current injection techniques
conventionally the PV inverter would disengage from the power grid by utilizing an anti-islanding technique.
Grid faults
Inverter topologies
Nevertheless, many countries have now implemented grid codes (GCs) to secure and regulate the operation of
the GCPV systems from exposure to grid faults. Modern advanced technologies have equipped the PV inverters
with the low-voltage ride-through (LVRT) capability to address the issue of grid faults and prevent equipment
failure, as it develops compliance of the GCPV systems with the GCs. This paper presents an overview of inverter
topologies which include the power processing stage, transformer and transformerless inverter, multilevel in-
verter, soft and hard switching, bidirectional and hybrid inverters. Thereafter, current injection techniques with
LVRT control strategy are discussed and examined. Such techniques have been classified under four main ca-
tegories: the current compensation controls, reactive current injection (RCI), linear current controls and non-
linear current controls. In addition, a comparative analysis has been presented, by thoroughly discussing the
merits and demerits of these control schemes, based on the performance parameters of complexity, stability,
robustness and power quality assessment.
1. Introduction Therefore, many current injection techniques have been studied and
implemented in simulations and practical prototypes. Further, the
Increasing research in the field of renewable energy has not only techniques are employed in the LVRT applications focusing on the grid
reduced the reliance on fossil energy, but has also offered an econom- faults to overcome the flash of power outages by injecting some amount
ical and efficient alternate to cope up with soaring energy demands of active or reactive current to stabilize the grid voltage.
(Obi and Bass, 2016; Obeidat, 2018; Commission, 2017; Commission, The key contribution of this paper is to discuss the possible methods
2016). Particularly, the cost of solar PV systems has declined more than and controllers in the current injection techniques to cater to such
63% from the year 2010 to 2018 (Fu et al., 2018). The PV systems are problems in the GCPV systems and the grid voltage. Plus, each tech-
typically working in GCPV and off-grid photovoltaic (OGPV) opera- nique will be discussed on the working principle and the performance
tions. The GCPV systems are connected to the power grid while OGPV comparison is presented to render an aerial overview of the techniques.
remain stand-alone systems to satisfy an isolated power demand, for The objectives of the review paper are explained as follows:
instances, a hybrid and battery energy storage system (BESS) (Islam and
Mamun, 2017; Roberts et al., 2019). The GCPV systems are exposed to 1. To discuss different techniques and controllers in current injec-
unbalanced grid faults and exposed to negative impacts on the systems. tion for the LVRT and the grid faults from the other researchers.
Consequently, the power system efficiency, power quality and relia- 2. To compare the performance of each controller particularly fo-
bility become a major concern (Almeida et al., 2016; Al-Shetwi et al., cusing on the reliability, robustness, stability, dynamic response,
2019; Parvez et al., 2016; Huka et al., 2018; Sosa et al., 2016). and harmonic compensation capability.
⁎
Corresponding authors.
E-mail addresses: zulhilman5107@um.edu.my (Z. Hassan), jeyraj@um.edu.my (J. Selvaraj).
https://doi.org/10.1016/j.solener.2020.06.085
Received 12 November 2019; Received in revised form 21 May 2020; Accepted 23 June 2020
0038-092X/ © 2020 International Solar Energy Society. Published by Elsevier Ltd. All rights reserved.
Z. Hassan, et al. Solar Energy 207 (2020) 851–873
Nomenclature AARC Average active reactive control

BC Backstepping control
Id Active current BCC Brake chopper circuits
c Center of the peak of FLC BESS Battery energy storage system
ud Controller for active current BPSC Balanced positive-sequence control
uq Controller for reactive current CAP Constant active power
Active current ratio CAC Constant active current
G Active current reference CARP Constant active-reactive power
P Active power CPC Constant peak current
Vg [n] Average grid voltage CRP Constant reactive power
Vo_avg [n] Average inverter output voltage CSI Current-source inverter
Q Constant component of reactive power DB Dead-beat
Gc (s ) Controller in control system DC Direct current
I Current DCL Dynamic current limitation
I (s ) Current in control loop DLG Double line to ground
i Current vector DSP Digital signal processor
E (s ) Disturbance in control loop DVCC Dual vector current controller
Gf (s ) Filter in control system DVR Dynamic voltage restorer
V Grid voltage RMS value ESS Energy storage system
Pg Grid active power FCL Fault current limiter
Qg Grid reactive power FLC Fuzzy logic control
L Inductance FLS Feedback linearization strategy
kI Integral gain FRT Fault ride-through
ri Integer of FLS GC Grid code
LT Interfacing inductance GCPV Grid-connected photovoltaic
RT Interfacing resistance HB Hysteresis band
p Instantaneous active power HC Harmonic compensation
g Instantaneous conductance IARC Instantaneous active reactive control
b Instantaneous susceptance LL Line to line
0 Line grid frequency LVRT Low-voltage ride-through
Lf Order of Lie derivatives MCAC Maximum current amplitude control
v Orthogonal voltage vector MCC Maximum current control
p Oscillatory active power MIC Modified inverter controller
q Oscillatory term of reactive power MLI Multilevel inverter
pu Per unit, ratio of actual to nominal voltage MPPT Maximum power point tracking
ipv Photovoltaic current OGPV Off-grid photovoltaic
kp Proportional gain PC Predictive controller
PWM delay in control system PCC Point of common coupling
In Rated current PCLC Peak current limit controller
Iq Reactive current PI Proportional integral
B Reactive current reference PNS Positive- and negative-sequence
q Reactive power PNSC Positive- and negative-sequence compensation
Kq Reactive power ratio PR Proportional resonant
R Resistance PSO Particle swarm optimization
Resonant frequency PV Photovoltaic
Ts Sample time delay RCI Reactive current injection
sd Sliding mode for active current RC Resonant control
sq Sliding mode for reactive current RMS Root mean square
Sn Switch signal of phases SCESS Super capacitor energy storage system
V Voltage SDBR Series dynamic breaking resistor
VPCC Voltage at the point of common coupling SLG Single line to ground
Vg Voltage fault terminal SMC Sliding mode control
v Voltage vector SRF Synchronous reference frame
Width of the bell curve of FLC STATCOM Static synchronous compensator
THD Total harmonic distortion
Abbreviations VAC Variable active current
VCFF Vector current controller with feedforward
ADL Adaptive DC-link VSI Voltage-source inverter
APRC Active power ripple cancellation
AC Alternating current
852
3. To present the guide by displaying the merits and demerits for breaking chopper and fault current limiter (FCL). While for the con-
other researchers to select an appropriate techniques to be im- troller based methods are emphasizing the modified inverter controller
plemented in the research works. (MIC), computational methods and other complementary methods. By
considering the merits, demerits, cost, complexity and fulfillment of
This paper is structured as follows: Section 2 study on the state-of- GCs, the authors conclude that the MIC method is the best among the
the-art of the previous research works and the differences are pre- other methods, followed by the FACTS and ESS method. The MIC
sented. Section 3 discusses the compliance of the grid codes (GCs) and method is able to retain the inverter connection with obligation of the
the policies implemented in several countries. Section 4 introduces the system safety and by injecting the active and reactive power control to
balanced and unbalanced grid faults by categorizing them under sym- provide grid support during FRT.
metrical and unsymmetrical faults. Meanwhile, Section 5 presents a The approaches proposed in (Parvez et al., 2016) are based on the
brief explanation of the LVRT control strategy while maintaining current control techniques which constitute of the linear and non-linear
compliance with the standardized GCs. Subsequently, Section 6 pre- methods. The linear methods comprise of proportional-integral (PI),
sents an overview of the inverter topologies which include the power proportional-resonant (PR) and repetitive current (RC) controllers.
processing stage, transformer and transformerless inverter, multilevel Meanwhile, the non-linear methods consist of dead-beat (DB), hyster-
inverter, soft and hard switching, bidirectional and hybrid inverters. esis and predictive controllers. The authors have discuss further on the
Following after, Section 7 discusses the current injection techniques analysis of the methods, the power quality issues related to the applied
under LVRT conditions listing several techniques, namely: the current methods, voltage and frequency issues, grid synchronization technique
compensation controls, reactive current injection (RCI), linear current and the performance for each explained method. The PR controller can
controls and non-linear current controls. The performance comparison be seen as the best solution due to its high dynamic performance and
of current controllers in terms of robustness, reliability, stability, dy- harmonic compensation capability to execute the LVRT during the grid
namic response and harmonic compensation capability are summarized faults occurrence.
in Section 8. Key challenges and future recommendations are included Another control scheme based on current control strategies are in-
in Section 9. Finally, the conclusion of the paper is carried out in troduced in Tian et al. (2014)) which incorporate the PI, PR, DB and RC
Section 10. controllers under dynamic performance methods. Meanwhile, the au-
thors have comprehensively explain on the controllers under symme-
trical and unsymmetrical faults which are the VCCF and DVVC con-
2. State-of-the-art trollers. The VCCF and DVVC are considered as the sequence separation
method that have the abilities to compensate the unbalanced voltage
As presented by Al-Shetwi et al. (2019); the authors are categorizing effectively and to improves the voltage in DC-link with a constant re-
the LVRT or also known as fault ride-through (FRT) enhancement active power injection, respectively.
methods into a group of the control type and connection configuration. To investigate the potential impact of voltage source inverter (VSI)
These two groups will highlight on the external devices-based methods under unbalanced faults, the authors in Jia et al. (2018) have in-
and modified controller-based methods. The external devices based vestigated the current control strategies for short circuit power
methods are focusing on FACTS devices, energy storage system (ESS),
Table 1
Summary of the state-of-the-art.
Author(s) Year of publication List of methods No. of methods Merits/Demerits
(Hassan et al. this paper) 2020 • PNSC • PR control 25 Yes

• PCLC • RC control
• ARPC • SFC
• IARC • DB control
• AARC • HC
• BPSC • PC
• CAP • SMC
• CRP • FLS
• CAC • BC
• VAC • FLC
• CPC • ANN
• MCC • GA
• PIBESScontrol
(Al-Shetwi et al., 2019) 2019 • SCESS • MIC 13 Yes
• BCC • PSO
• STATCOM • FLC
• DVR • DCL
• FCL • FLS
• SDBR • ADL
• PI controller
(Parvez et al., 2016) 2016 • PR controller • DB controller 6 Yes
• RC controller • Hysteresis controller
• Droop controller • Predictive controller
(Tian et al., 2014) 2014 • PI controller • RC controller 8 No
• PR controller • VCCF
• DB controller • DVCC
• Instantaneous active power • MCAC
(Jia et al., 2018) 2017 • Instantaneous reactive power • CRP 9 No
• Balanced current control • CARP
• Oscillating power control • Semi-flexible PNSC
• CAP • Flexible PNSC
•
853
provision by classifying into two groups, namely power-characteristic- impedance and short circuit faults. Moreover, a surge current flows due
oriented and voltage-support-oriented control strategy. The authors to the grid faults, here, a robust protective mechanism is required to
have emphasized the important of controller selection under un- prevent severe damage to the equipment. There are several types of grid
balanced faults based on the purpose and desired objective. The case faults, namely as:
studies have been proposed to illustrate the disparity among the control
strategies by considering the voltage and current limits while injecting • Line to line (LL)
the active and reactive power. • Single line to ground (SLG)
On the other hand, this paper proposes a comprehensive review on • Double line to ground (DLG)
current injection techniques for LVRT and grid fault conditions in the • Open conductor fault
GCPV systems by covering the mentioned methods above and im- • Balanced three-phase fault
proving with another techniques and controllers. These controllers
define different LVRT functions and services with a comprehensive Symmetrical fault and unsymmetrical fault are the two methods to
discussion on the current compensation controls, RCI controls, linear analyze the operational balanced and unbalanced faults (Patel, 2018;
and nonlinear current controls. Table 1 exhibits the discussed review Miret et al., 2013; Lin et al., 2018). Where, symmetrical (balanced)
papers in this section by tabulating the list of methods and comparison fault is a fault when all the three-phase voltage drops are of the same
in merits and demerits among the proposed controllers. magnitude proportional to the severity of the grid faults. The common
type associated to this fault is the balanced three-phase fault. By con-
trast, unsymmetrical fault constitutes severe unbalanced operating
3. Compliance of grid codes (GCs)
conditions, which involve only one or two phases when the phase-
voltage drop is uneven depending upon the nature of the fault. Here,
Several countries like Germany, Romania, U.S.A, China, South
the associated types of grid faults remain SLG, LL and DLG (Byung-Ik
Africa and Malaysia have employed GCs as summarized in Table 2 to
et al., 2012; Hyo-Sang et al., 2010; Sadeghkhani et al., 2018).
cater to the grid faults in GCPV systems (E. Commission, 2017, 2016;
Cabrera-Tobar et al., 2016; Dhimish et al., 2017a, 2017b; Luo, 2018).
There are governing bodies that regulate the policy, for instance, En-
5. Low-voltage ride-through (LVRT)
ergy Commission (EC) in Malaysia who regulate the grid codes for the
Peninsular Malaysia, Sabah and Labuan. The policy covers the grid
Conventionally, a sudden stoppage of the GCPV system, due to se-
codes for the power interconnection from generating plants that con-
vere grid faults, would trigger the islanding protection and shut down
nected to the power grid system. These plants include the power gen-
the inverter causing power outages, voltage flickers and energy losses
eration from conventional and renewable energy power plants.
(Yang et al., 2014b; Papanikolaou, 2013; Ramirez et al., 2017).
Conventionally, anti-islanding techniques were employed to disen-
Therefore, the LVRT technique had been proposed by researchers to
gage the PV inverters during fault conditions. Here, the main concern
control and regulate the issue of grid faults and to comply with the GCs.
was the spike of DC voltage exceeding the over-voltage limit. Therefore,
Fig. 1 shows the basic concept of LVRT and anti-islanding require-
researchers have proposed the LVRT in order to ensure a reliable per-
ments, where it describes that during the event of grid faults, the PV
formance of the GCPV without disengaging the system from the grid,
system should remain connected to the power grid if the voltage at the
during fault conditions, and maintaining compliance with the GCs
point of common coupling (PCC) stayed above the line and contra-
(Nanou and Papathanassiou, 2014; Weise, 2015; Garnica Lopez et al.,
riwise. The PV system should not be disconnected from the power grid
2018; Seung-Tak et al., 2013). The key element of this study is to ex-
for at least time t1 when the voltage at the PCC (VPCC ) drops to 0% or
plore the balanced and unbalanced grid faults, while highlighting the
lower than V0 pu (per unit, the ratio of actual to nominal voltage) (Al-
LVRT control strategy, with compliance to the standardized GCs, as a
Shetwi et al., 2018). The figure also interpreting the short-term is-
solution to confront and overcome such shortcomings.
landing within t1 to t3 with a specific value of the VPCC to be followed. If
the VPCC remains lower than the V1 for the period of t3 to t5, thus the anti-
4. Balanced and unbalanced grid faults categories islanding protection will be triggered and the PV system is obliged to
disconnect from the power grid.
Generally, the GCPV system operates under balanced condition with For instance, Germany determine the PV system should remain
ideal load currents and voltages, working within the inverter opera- connected with the power grid at least for 0.15 secs when the VPCC
tional limits. The occurrence of grid fault causes an unbalanced state for drops to 0 pu and should recover the PV system within 1.5 secs with the
the GCPV system. This is due to natural interference, exposure to VPCC settles at 0.9 pu. In Peninsular Malaysia, it follows the same con-
technical errors or connection of large number of loads to the power figuration as Germany, but different for configuration in Sabah and
grid (Ammiche et al., 2018; Farhang et al., 2015). Labuan for the recovery time after the faults. The difference is de-
A grid fault occurs when there is an interruption in current flow of pending on the condition for the grid at the low-voltage or distribution
the inverter within few milliseconds and as a consequence it creates low level, the grid usage, system services, grid expansion and general grid
Table 2
Grid codes policy in several countries (E. Commission, 2017, 2016; Cabrera-Tobar et al., 2016).
Country Organization Policy title Version
Germany BDEW Power plants connected to the medium-voltage grid 2018

Generating plants connected to the medium-voltage network
Romania ANRE Technical conditions for connection to public electricity networks for PV power plants 2013
U.S.A. PREPA Technical requirements for interconnecting wind and PV generation 2012
FERC LGIA Standard large generator agreement 2006
China NEA Generating plants connected to the medium-voltage network 2012
South Africa NERSA Grid connection code for renewable power plants connected to the electricity transmission system or the distribution system in South 2012
Africa
Malaysia EC Grid code for Peninsular Malaysia 2016
EC Grid code for Sabah and Labuan 2017
854
power processing stage, transformer and transformerless inverter,

multilevel inverter, soft and hard switching inverter, bidirectional and
hybrid inverter as delineated in Fig. 2. The following subsection ela-
borates each category from the inverter topologies as reported in Zeb
(2018), Sajedi et al. (2019), Prabaharan and Palanisamy (2017), Kala
and Arora (2017), Jana et al. (2017), Dogga and Pathak (2019), Babaie
and Asl (2016), Ahmad and Singh (2017), Chakraborty et al. (2017)
and Islam et al. (2015).
6.1. Power processing stage
This inverter topology consists of two main types, namely as single-

stage and multiple stage inverter. The single-stage inverter can execute
various tasks and functions, for instance, the control of current injection
into the grid during fault occurrence, the voltage amplification and the
control of maximum power point tracking (MPPT). The latest tech-
Fig. 1. Basic concept of LVRT and anti-islanding requirements. nology has implemented the high-frequency transformer or the trans-
formerless converter to replace the line frequency transformer which
Table 3 needs to cope with the burden of weight and efficiency losses of 2%.
LVRT requirements in grid codes (GCs) (E. Commission, 2017, 2016; Cabrera- However, the single-stage inverter has several drawbacks such as the
Tobar et al., 2016; Lin et al., 2018; Jerin et al., 2018). inability to deal with a low range of input voltage, low power quality
and reduced power capacity.
Country During fault After fault
Multiple-stage inverter imposed two or more stages of voltage
V0 (pu) t1 (pu) V2 (pu) t2 (pu) conversion, typically the first stage will convert the DC to DC voltage
and the second stage will convert the DC to AC voltage. In the past
Germany 0 0.150 0.90 1.50
decade, transformerless inverter is equipped with this topology and
Romania 0 0.625 0.90 3.00
U.S.A. (Puerto Rico) 0.15 0.600 0.85 3.00 works greatly to operate in the low DC voltage. Nonetheless, the un-
China 0 0.150 0.90 2.000 desired transient occurs in the operation of high switching frequency,
South Africa 0 0.150 0.85 2.00 thus the DC-link capacitor is attached at the inverter input to filter the
Ireland/Canada 0.15 0.625 0.90 3.00 transient from flowing backwards. The drawbacks of the DC-link are the
Italy 0.20 0.500 0.90 2.00
bulky size, weight, lifetime and cost. Fig. 3 shows the single and mul-
Denmark 0.25 0.150 0.75 0.75
Spain 0.20 0.500 0.95 1.50 tiple-stage inverter with the DC-link capacitor.
Malaysia (Peninsular) 0 0.150 0.90 1.50
Malaysia (Sabah & Labuan) 0 0.150 0.90 1.65 6.2. Transformer and transformerless inverter
The transformerless inverter is the most acknowledged inverter in

operation. Table 3 shows the LVRT requirements in several countries
the modern era of the GCPV systems. The merits are contributed by its
related to the illustrated figure in Fig. 1, where each country must obey
high efficiency, cost-effective, lighter and compact. However, the in-
and follow the time frame and voltage limit during and after the faults
verter facing the issues with the DC current injection which require
occurred (Shin et al., 2015; Carrasco et al., 2013).
extra circuitry. Secondly, there is no galvanic isolation between the grid
and the PV array which can cause voltage fluctuation and worsen the
6. Inverter topologies for current controls grid faults. The differences between the transformer and transformer-
less inverters are simplified in Table 4.
An Inverter can be classified into four types which are the cen-
tralized inverters, string inverters, multi-string inverters and AC cell or 6.3. Multilevel inverter
AC module inverters. These four types will apply one or more inverter
topologies. There are various inverter topologies reported in the re- Multilevel inverter has become a common topology in inverter
search field. In this section, classification of the inverter topologies are technology today. The topology offers numerous advantages where the
presented in six divisions based on different configurations, namely, the AC voltage output is smoother with staircase-shape sinusoidal
Fig. 2. Classification of inverter topologies.
855
Fig. 3. (a) The single-stage and (b) multiple-stage inverter with DC-link capacitor (Zeb, 2018).
waveform resulting from the multilevel features. Other than that, the 6.6. Hybrid inverter
harmonic distortion is significantly reduced and the filter components is
decreased. There are three multilevel inverter topologies presented in Many hybrid inverter have been proposed in recent years to com-
this section which are the half-bridge diode clamped inverters, full- pete with the high demand and requirement of grid codes for the LVRT
bridge single leg clamped inverters and cascaded inverters as shown in feature and to cope with the power quality issue. This topology also will
Fig. 4. increase the power output efficiency, hence will reduce the components
and cost. The versatility of the topology have achieved many research
6.4. Soft and hard switching inverter findings to be implemented in industry, for example, the developed
cascaded MLI, cascaded sub-MLI, multilevel DC-link inverter, hybrid
Soft-switching inverter employed a high-frequency switching and multi-cell converter, series diode clamped H-bridge MLI and switched
has superior benefits compared to the hard-switching inverter. High series/parallel source MLI for the LV topologies. Meanwhile, the MV
power density, high efficiency, better performance are the apparent and HV topologies consist of cascaded basic blocks MLI, cascaded
advantages of soft-switching over the hard-switching inverter. Despite modified H-bridge MLI, chain cell MLI, cross connected source based
that, the high-frequency switching has caused the EMI issues, switching MLI, level doubling network cascaded MLI and switched capacitor cell
losses, current and voltage spikes due to the stray inductance and hybrid MLI.
parasitic capacitance. Hence, the resonant soft switching technique is
applied to ensure the voltage across or the current through the
7. Current injection techniques under LVRT conditions
switching devices to be zero. Zero-voltage switching and zero-current
switching are the two notable types under soft-switching topology to
The current injection techniques can be classified into four cate-
reduce dv/dt and di/dt, respectively. Fig. 5 depicts the LCLC resonant
gories namely: current compensation controls, reactive current injec-
bidirectional DC-DC converter integrates with non-isolated buck boost,
tion (RCI), linear current controls and non-linear current controls as
soft-switching and zero-voltage switching features.
depicted in Fig. 6. The main objective to implement these controllers
remains to control the power grid with current injection techniques
6.5. Bidirectional inverter during the grid faults. These techniques have applied the injection of
active and reactive currents but differ in controller strategies as sum-
As the name implies, a bidirectional inverter is a topology used to marized in Table 5.
regulate and monitor the power flow between a DC bus and an AC grid.
The topology is also capable to restrict the voltage expanse to a certain
permissible range of voltages. Several merits are noticed for this to- 7.1. Current compensation controls
pology such as the capability to perform multiple conversion between
DC-DC, DC-AC and AC-DC. The flexibility to integrate with other Six current injection techniques that cater to the issue of grid faults
topologies is another merit point for bidirectional inverter as shown in during LVRT conditions have been listed. The unbalanced (un-
Fig. 9. Many sub-topologies are reported by researchers, for instance, symmetrical) grid fault voltage would produce a significant double line
the buck-boost DC-DC converter, quasi Z-source inverter, dual active grid frequency, 2 0 power ripples in the DC-link capacitors. The dif-
bridge (DAB) inverter and bridgeless boost. ferent sequence of currents and voltages in active and reactive power
Integration with isolated and non-isolated buck-boost inverter de- causes the exchange of the controls. Therefore, the current compensa-
pends on the application of high power and performance of the effi- tion controls are required to resolve the issues. These techniques are
ciency and simplicity. The quasi Z-source inverter combines the fea- implemented to estimate the amount of the active or reactive current
tures of the boost and VSI into a single-stage inverter, resulting in injection to the grid by power converter during LVRT conditions. Such
higher efficiency compared to the conventional boost and VSI that built current compensation controls are presented by researchers in
separately. This is due to the compact structure, less harmonic and less (Saccomando and Svensson, 2001; Teodorescu et al., 2011; Hunter
distortion (Zhou et al., 2012). Meanwhile, the DAB inverter topology is et al., 2016; Chen et al., 2016a, 2016b; Lee et al., 2011; Alepuz, 2009;
typically employed for the single-stage isolated bidirectional inverter RodrÍguez et al., 2009; RodrÍguez et al., 2006; Chou et al., 2011;
which offers a bidirectional power flow, active power flow control, soft Magueed et al., 2004). Eq. (1) formulates the instantaneous active
switching and isolation by a high-frequency transformer (Cho et al., power, p , generated by the three-phase power converter of GCPV
2016). system as:
Table 4
Differences between transformer and transformerless inverter (Jana et al., 2017).
Inverter Merits Demerits
Line-frequency transformer Easy design, high reliability, high safety due to galvanic isolation Low efficiency, heavy and bulky
High-frequency transformer Easy design, high efficiency, high safety due to galvanic isolation Complex design, costly technology
Transformerless Compact, lightweight, high efficiency, cost-effective Complex design,
856
Fig. 4. Multilevel inverter topologies, (a) three level half-bridge diode clamped inverter, (b) full-bridge single leg clamped inverter and (c) cascaded inverters (Zeb,
2018; Jana et al., 2017).
Fig. 5. Non-isolated buck boost LCLC resonant bidirectional DC-DC converter (Wang et al., 2019).
Fig. 6. Categorization of current injection techniques under LVRT conditions for GCPV system.
p = v· i (1) current, Eq. (1) is further elaborated as follows:
where p is the instantaneous active power, v is the voltage vector in the

p = (v+ + v )·(i+ + i ) (2)
PCC, v = (va, vb, vc ) and i is the injected current vector, i = (ia, ib, ic ) . The interaction between the reactive currents and the voltages
Considering symmetrical (balanced) condition of the voltage and the generates the reactive current injection as expressed in Eq. (3) and Eq.
857
Table 5
Summary of current injection techniques with the controller strategy.
Current injection techniques Controller strategy
Current compensation controls Sequences control of voltage and current

Reactive current injection (RCI) Constant, variable and maximum current
controls
Linear current controls Frame control of dq, andabc
Non-linear current controls Frame control ofabc
Fig. 8. Block diagram of DVCC that imposed positive and negative SRF
(Saccomando and Svensson, 2001).
(4) which constitute by the form of positive–negative-sequence (PNS) of
the voltage and current.
b. Dual Vector Current Controller (DVCC)
p = (v+ + v )·(i+ + i ) (3)
p = (v+· i+ + v ·i ) + (v+· i + v · i+) = P + p~ (4) The DVCC method as depicted in Fig. 8 injects both of the PNS currents
in order to minimize the ripple power in the converter. This may result
where v+ + v are the PNS voltage components, i+ + i are the PNS in high peak current and initiate the overcurrent protection of the
current components, P is active power and p~ is the oscillatory active converter. The double line grid frequency, 2 0 power ripples can be
power. Eq. (5) expresses the reactive power, q by taking the funda- reduced by using two methods which are DVCC and IARC. The 2 0
mental of active power, p in Eq. (1). The reactive power, q is produced power ripples are reduced by injecting negative-sequence current or
by the power converter. The interaction of v andi are expressed in the harmonic current. Nevertheless, these two methods will ensure a high
form of cross product to generate the reactive power, q . peak current in the inverter AC output during LVRT condition.
Fundamentally the working of the PNSC can be described as the
q = |v × iq| = v iq (5)
controller that is associated with the PNS components in order to cal-
where q is the reactive power and v is the orthogonal version of 90° culate their reference current vectors. Furthermore, the PNSC allows
leaded of the voltage vector in the PCC, v . Eq. (7) has two different parts the active and reactive instantaneous power to use both synchronous
which are the constant component, Q and oscillatory term, q . Hence, and stationary reference frames to cater to the required power at-
the reactive power for the PNS reactive current is describe in the fol- tenuation level. Eqs. (8) and (9) can be utilized to calculate the constant
lowing equation: components of active power, P and reactive power, Q respectively,
where the PNS components at the frequency, iq +, iq replace the vector
q = (v+ + v )·(i+ + i ) (6)
component of the current, i+ + i in Eq. (4).
q = (v+· i+ + v ·i )+(v+·i + v · i+) = Q + q~ (7) P = v+·ip + + v ·ip ; 0 = v+·ip + v · ip + (8)
where Q is the constant component of reactive power and q~ is the os- Q = v+· iq + + v · iq ; 0 = v+· iq + v · iq + (9)
cillatory term of reactive power.
where are the PNS components at the fundamental frequency
ip,+q, ip, q
7.1.1. Positive-Negative-Sequence Compensation (PNSC) respectively. Hence, the active and reactive current vectors are calcu-
There are two types of controllers proposed in the PNSC lated as:
(Teodorescu et al., 2011) namely the VCCF and DVVC (Saccomando P
ip' = g± (v+ v ); g± =
and Svensson, 2001; Alepuz, 2009; Chou et al., 2011) that have been |v+|2 |v |2 (10)
typically used to inject the required PNS reactive current as explained
Q
below: iq' = b± (v+ v ); b± =
|v+|2 |v |2 (11)
a. Vector Current Controller with Feedforward (VCCF)
Several current controllers have been proposed to counter the main

issue of the inverter and rectifier operation to minimize the DC-link
power ripples. Thus, the current controllers use the phase-sequence
components to deal with the unbalance grid faults. Synchronous re-
ference frame (SRF) is one of the current controllers proposed in the
VCCF. This method only produces the positive-sequence reactive cur-
rent, as a proper controller can reduce the probability of high current in
the controller. The VCCF architecture is shown in Fig. 7.
Fig. 7. Block diagram of VCCF that uses positive SRF (Saccomando and
Svensson, 2001).
Fig. 9. Euclidean plane for IARC and PNSC current vector for positive- and
negative sequence (RodrÍguez et al., 2006).
858
Finally, by adding Eqs. (10) and (11), the final reference current is the grid fault, but offers large harmonics and in turn produces un-
expressed as: balanced currents (Zeb, 2018). To resolve the issues, the fundamental
equations in Eqs. (1) and (5) are imposed in the IARC equations in Eqs.
i' = ip' + iq' = g± (v+ v ) + b± (v+ v ) (12) (18) and (19) which are stated as:
where g is the instantaneous conductance and b is the instantaneous
susceptance. • Alignment between current and voltage vectors, v will increase the
active power
7.1.2. Peak current limit controller (PCLC) • Alignment between current and orthogonal voltage vectors, v will
This controller has the capability to remove the negative-sequence generate the reactive power
from power grid by injecting the active, reactive or inductive negative-
sequence current. Eq. (13) is formulated for the maximum amplitude of The following equations are derived to determine the reference
the negative-sequence current injection, IPCLC
n
to ensure that the max- currents:
imum current operates within the predetermined value of Imax . The P Q
predetermined value of Imax evaluates several factors, for instances, the ip = v ; iq = 2 v
|v|2 |v| (18)
range value of overcurrent protection, the capacity and stress limit of
the power transistor and the system stability. Hence, the final reference current is:
n 4 2 4 2
i = ip + iq (19)
IPCLC = I pcos +k + Ip cos 2 +k 1 + Imax
3 3 (13)
where ip , iq are the active and reactive current vectors, respectively. Eq.
where Imax is the predetermined value of the maximum current and k is (19) shows the final active and reactive reference currents that can
determined as the following conditions: efficiently inject the currents during the grid faults. Fig. 9 illustrated the
IARC and PNSC current compensation controls on the Euclidean plane.
0, 3
< 3 ‘Equipower line’ is used to set a boundary for the amplitude which is
k = 1, < , and = p + n + 1 2
perpendicular to v line. The oscillatory of reactive power, q is pro-
3
5 portionate to the current vector over equipower line. The AARC and
1, < 3 BPSC, that will be explained later, have the same direction of the cur-
rent vectors as the IARC. However, both methods are generating active
power oscillation, hence the amplitudes are not instantaneously con-
7.1.3. Active power ripple cancellation (APRC)
trolled.
The predefined value of the maximum current, Imax will determine
the maximum limit allowed for active or reactive current injection. The
APRC is imposed to reduce the active power ripple without exceeding 7.1.5. Average active reactive control (AARC)
the predefinedImax . Eq. (14) indicates the instantaneous active power During the grid faults, the inadvertent active power flows in the
dealing with the AC-side capacitive filter of the converter. inverter generate the reactive power oscillation. The AARC employs a
similar approach to the PNSC strategy in order to eliminate this issue.
3 s s
po (t ) = (vq iq + vds ids ) The reference currents, ip , iq are proportional to the orthogonal voltage
2 vector, v . Therefore, the AARC strategy is imposed to calculate the
3 p p instantaneous conductance and susceptance to dictate the reference for
= {V I cos( p) + V nI ncos( n)}
2 the active and reactive current vectorsip , iq . Eq. (20) expresses the
condition under unbalanced grid faults where the module |v|2 contains
+ V pI n cos(2 t + 1 + 2 + n) the grid frequency oscillation.
+ V nI pcos(2 t + 1 + p + 2)} (14) |v|2 = |v+|2 + |v |2 + 2 |v+||v | cos(2 t + + ) (20)
where are the PNS components of grid voltage, respectively,
V p, Vn
The cancellation of high-order current harmonics can be determined
I p, I n are the PNS magnitude of the current converter,
using Eqs. (21) and (22) as follows:
1, 2 , n represent phase angle of the PNS voltages with respect to the
reference axis, t , p, n are the phase angle of the PNS current, re- P
ip = Gv ; G =
spectively and vqs iqs , vds ids are the DC components for reactive current. V2 (21)
The magnitude of the ripple component of po (t ) is stated in Eq. (15):
3 Q
| |pcos |2 +|psin |2 iq = Bv; B =
p| = V2 (22)
2 (15)
where |pcos | = V pI ncos( 1 + 2 n p
n ) + V I cos( 1 + 2 + p ) and where V is the grid voltage RMS value. In the AARC strategy, G and B
|psin | = V pI nsin( 1 + 2 n ) + V I sin( 1 + 2 + p) . for active and reactive current references, respectively, are considered
n p
Finally, Eq. (17) expresses the corresponding amplitude of the po- as constant values. Thus, the collective RMS value can be expressed as
sitive-sequence current by combining Eq. (16) (active power ripple Eq. (23).
cancellation) with Eq. (13) from PCLC method.
1 T
Vn p V = |v|2 dt = |v+|2 + |v |2
In = ·I T 0 (23)
Vp (16)
Eq. (24) substitute and V with Eqs. (20) and (23) respectively,
|v|2
Imax
Icpp = where the sum product of the average value, P and the oscillating term,
( ) Vn 2
Vp
+2 ( ) cos (
Vn
Vp
+k
4
3 )+1 (17)
p~ as formulated in Eq. (24) will calculate and justify the immediate
active power injection to the unbalanced grid faults.
|v|2
7.1.4. Instantaneous active reactive control (IARC) p = ip · v = P = P + p~
The IARC produces the active and reactive power constant during V2 (24)
859
7.1.6. Balanced Positive-Sequence control (BPSC) minimum reactive current as stated in Eq. (29). Eq. (31) expresses the
The BPSC is considered as the simplest strategy in current com- equation for CRP control:
pensation control for the active reactive current control under grid fault
Kq
conditions. This controller aims to maintain balance of the three-phase Iq =
sinusoidal currents while injecting them into the grid with positive-
Vg (31)
sequence component only. As a result, this controller is able to generate where K q is the reactive power ratio. Fig. 12 illustrate the relationship
a lower peak current for a particular average active and reactive current between reactive current, Iq and reactive power ratio, K q as elaborated
transfer during grid faults. in Eqs. (29) and (31) respectively. The solid-straight line represents Eq.
Eqs. (25) and (26) determine the BPSC under a perfect balanced (29) under ideal conditions, while the other three curve-dotted lines
positive-sequence sinusoidal waveform. By excluding the negative-se- represent the Eq. (31). Based on the Eq. (29), all Iq must be above the
quence in Eqs. (10) and (11), the BPSC will only injects the positive- solid-straight line in order to fulfil the set condition. Three values of K q
sequence as expressed in the following equations: (0.7, 0.5 and 0.3) have been tested in Eq. (31) but only K q of 0.7 and 0.5
P pass, while K q of 0.3 infringe the condition in Eq. (29).
ip = G+v+; G+ =
|v+|2 (25)
7.2.3. Constant active current (CAC)
Q
iq' = B+v+; B+ = This CAC controller has the capability to inject a constant active
|v+|2 (26)
current under unbalanced conditions with current reference. Relate to
Eqs. (27) and (28) are the active and reactive power equations for Eq. (29), the active current, Id has to keep the active current constant at
the BPSC controller to resolve the power oscillations ( p and q ) during a specific period of time during the grid fault. It can be any value from 0
the grid faults due to the response between PNS components which to 1 as the ratio can be stated as below:
consists of the current and voltage sequences. Besides, the BPSC is the Id = In 1 (32)
only current compensation control that allows the acquisition of both
sinusoidal and balanced currents. where is defined as the active current ratio.
p = v· ip = (v+· ip ) + (v · ip ) = P + p~ (27)
7.2.4. Variable active current (VAC)
q = v ·ip = (v ·iq ) + (v · iq ) = Q + q~
+ The VAC controller applies the equation provided in Eq. (33) where
(28)
the reactive current, Iq is firstly derived in Eq. (29) and the variable
active current, Id based on the reactive current, Iq and maximum cur-
7.2. Reactive current injection (RCI) rent, Imax
Id = Imax Iq (33)
Neumann and Erlich (2012) have received considerable attention by
explaining on the RCI techniques to provide voltage support during the where Imax is the predetermined maximum current for overcurrent
LVRT condition. Fig. 10 shows the step response with predetermined protection of the inverter.
rise time and settling time of 30 ms and 60 ms respectively, while
the limit of tolerance band is set between −10% to +20% from the 7.2.5. Constant peak current (CPC)
current reference. Firstly, the rise time is applied for the reactive cur- The CPC controller is utilized to provide the low and constant peak
rent to reach the tolerance band, then the settling time is used for the current injection in the inverter during grid faults. As a result, the in-
reactive current to oscillate within the tolerance band. verter remains connected, as it is not affected by islanding or over-
Fig. 11 depicts the grid code requirements for the RCI during LVRT current protection. The CPC in the dq -frame is expressed in Eq. (34) and
condition to ensure the voltage support and applied various controllers (35).
to satisfy the grid code requirements. Therefore, this section will review
the RCI techniques that have been explained in Al-Shetwi et al. (2019), Id = n2 k 2 (1 Vg ) 2 In
0.5p . u Vg < 0.9p . u;
Jia et al. (2018), Teodorescu et al. (2011), Miret et al. (2015), Yang Iq = k (1 Vg ) In (34)
et al. (2014a), Gonçalves et al. (2016), Elyaalaoui et al. (2019),
Brandao et al. (2019) and Oon et al. (2018).
Id = n2 1 In
Vg < 0.5p . u ;
Iq k (1 Vg ) (29) Iq = In (35)
where Iq is the reactive current on the y-axis, Vg is the PV inverter
terminal voltage on the x-axis and k is droop k = 2 .
7.2.1. Constant active power (CAP)

The objective of CAP controller is to nullify the output active power
oscillation and to sustain the active power injection during grid fault.
Hence, the average CAP remains constant during this short period of
time. The active current in the CAP controller is expressed as:
Kp
Id = In
Vg (30)
where Id is defined as the active current, Kp is the active power ratio, Vg

is the voltage fault terminal (pu) and In is the rated current.
7.2.2. Constant reactive power (CRP)

In contrast with CAP, CRP utilizes the reactive current references to
maintain the reactive power during the grid faults. Nevertheless, the Fig. 10. Step response of the RCI during LVRT condition (Neumann and Erlich,
controller has to ensure the reactive power fulfils the requirement of the 2012).
860
control, PR-based control and resonant control (RC). There are three
frame controls, also known as stationary reference frame, namely, the
dq, and abc frame controls as classified in Table 6. The linear current
controllers have referred to the previous work done in (Parvez et al.,
2016; Teodorescu et al., 2011; Razali et al., 2014; Meo and Perfetto,
2002; Hassaine et al., 2014; Blaabjerg et al., 2006; Timbus et al., 2009;
Karanayil et al., 2014; Yang et al., 2013; Prasad et al., 2016).
7.3.1. Proportional-Integral (PI)-based control

PI-based control typically used in current-source inverters (CSI) are
affiliated with dq control due to the ability in controlling DC variables
as depicted in Fig. 13. This controller offers the flexibility to be utilized
in abc frame where it is able to transform the grid currents from abc into
the dq control frame that rotates simultaneously with the grid voltage.
This controller has its own drawbacks, as it is unable to identify a si-
nusoidal reference without steady-state error and it cannot eliminate
the interruption, typically caused by a periodic signal. The equations for
Fig. 11. Grid code requirement for RCI during LVRT condition (Elyaalaoui
the PI current controller and PI-dq control are shown in Eqs. (37) and
et al., 2019).
(38), respectively.
kI
GPI (s ) = kp +
s (37)
kI
kp + s
0
GPI (s )dq = kI
0 kp + s (38)
where kp,kI are the integral and proportional gains, respectively. Re-
lated to the LVRT conditions under unbalanced grid faults, there are
several methods to solve the harmonics from the inverse sequence in
the grid voltage, which are:
a. Addition of the PNS components without voltage feed-forward

(Fig. 14). It is able to remove the harmonics, however, it would
double the computational exertion.
b. Implementation of the cross-coupling terms and voltage feed-for-
ward.
7.3.2. Proportional-Resonant (PR)-based control

PR-based control is a compliment controller to the PI-based control
Fig. 12. Relationship between reactive current, Iq and reactive power ratio, K q
(Oon et al., 2018).
with the capability to identify the sinusoidal reference without steady-
state error in the frame as illustrated in Fig. 15. This controller can
be imposed in both abc and frame and also can transform from the
7.2.6. Maximum current control (MCC) existing abc to frame by using a transformation module. Eq. (39)
The fundamental of the MCC control is based on the reactive cur- shows the expression of PR controller in frame with the main ob-
rent, Iq in Eq. (29) and varies in derivation of the active current, Id as jective to attain a high gain in the resonance frequency spectrum.
formulated in Eq. (36). This controller aims to protect the inverter from
overcurrent damage. Here, the value of the maximum current, Imax kp +
ki s
0
s2 + 2
depends on the design architecture and controller of the inverter fo- GPR = ki s
cusing on the power electronic switches. 0 kp +
s2 + 2 (39)
Id = 2
Imax Iq2 (36) where is used for the resonance frequency.
7.3.3. Resonant control (RC)

7.3. Linear current controls The active and reactive power references generate the dq PNS
components in the RC method (also known as Repetitive Control).
This section will elaborate on the linear current controllers that There are several drawbacks in implementing the RC method which are
have been employed as the current injection techniques during the grid the failure in periodic signals tracking and the PNS components feed-
faults under LVRT conditions. The control techniques are the PI-based back errors. The RC method can be categorize into two methods to
Table 6
Classification of frame controls with the associated current controllers.
Frame control Transformation Description Current Controllers Equation
dq abc dq Converts the current and voltages into a rotating reference system in the grid PI-based Eq. (36)
abc Imposed an infinite gain at the resonant frequency to suppress the steady state error PR-based Eq. (37)
abc Separate controller for every grid current and proportional to the sampling frequency PI, PR, Hysteresis, DB and PC Eq. (40)
861
Fig. 13. Block diagram for PI-based control with dq -control (Parvez et al., 2016).
The synchronous controller, Gdq (s ) implements the inverse trans-

formation on the frame, where G (s ) to compensate the feedback
errors in the PNS components can be expressed as Eq. (41):
2kI c s
0
1 s2 + 2 c s + 2
G (s ) = 2kI c s
2 0
s2 + 2 c s + 2 (41)
7.3.4. State feedback control (SFC)

A linear state feedback control (SFC) is a simple, robust and high
reliable approach to be implemented in external disturbance rejection
in grid fault to compensate the current injection in the LVRT. The
controller can operate in stand-alone mode or integrate with another
controllers to form a hybrid controller, for instance, integration with
conventional PI controller and perturbation estimation. Fig. 17 shows
the SFC system which integrate with PI controller to maintain the
control signal from the PI controller and eliminate the need of feed-
forward compensation (Yang et al., 2017; Taveiros et al., 2015). The
SFC equation is written in Eq. (42) as follows:
x A + Bk Bke x 0 B d; y = [ x
= xa + 1 u + C 0 ] xa
xa C 0 0 (42)
where A, B and C are the system input and output matrices, respec-
tively. While x , u , y and d are the sate vector, input, output and dis-
turbance in the control system, respectively. The feedback gains are
denoted as k and ke .
Fig. 14. PI-based control with the PNS components (without voltage feed-for-
ward) (Teodorescu et al., 2011). 7.4. Non-Linear current controls
resolve the drawbacks as explained below: Generally, the non-linear current controllers will associate with the
abc frame control (Fig. 18) as it has the capability to control the grid
a. Internal model principle current with a separate controller. Thus, it will give more control
flexibility and usability. Such current controllers as the DB, hysteresis
To ensure that the RC can reject the unnecessary disturbance, the and PC require three or more independent controllers and employ the
model of the disturbance and the current reference should be involved matrix transfer function determined by Eq. (43). The abc frame control
in the closed-loop control feedback. The main objective is to counter the is preferable due to its high dynamic response and the compatibility
first drawback in periodic signals tracking. Eq. (40) is derived based on with the digital signal processor (DSP). This section will review several
the block diagram in Fig. 16. non-linear current controllers as elaborated in previous research works
in (Ramirez et al., 2017; Hassaine et al., 2014; Blaabjerg et al., 2006;
I (s ) (s 2 + 2)·(0.5Ts s + 1)
= Cortes et al., 2008; Mattavelli et al., 2005; Mattavelli et al., 2003;
(Ls + R)·(s + 2)·(0.5Ts s + 1) + (kp s 2 + k1 s + kp (40)
E (s ) 2 2)
Malesani et al., 1999; Chatterjee and Mohanty, 2016; Krishna et al.,
where I (s ) is the current in the control loop, E (s ) is the disturbance, L is 2010; Dai et al., 2009; Bode et al., 2005; Shukla et al., 2008; Yu and
the grid inductance and R is the grid resistance. Chang, 2005; Akagi et al., 2007; Galvez-Carrillo et al., 2009; Liu, 2013;
Meral and Celik, 2019; Abadlia et al., 2017; Eddine et al., 2016; Liu
b. Coordinate transformation et al., 2020; Soufi et al., 2016; Sekhar and Mishra, 2014; Montoya et al.,
2018; Lalili et al., 2011).
862
Fig. 15. Block diagram for PR-based control with -control (Parvez et al., 2016).
RT
T 1
1 e LT s z
abc 1 az 1
GDB (s ) = =
b (1 az 1) 1
RT
T
e LT s 1 (1 az 1)
RT
(44)
where RT is the equivalent interfacing resistance of the inverter, LT is
the equivalent interfacing inductance of the inverter and Ts is the
sample time delay.
Fig. 16. Block diagram for RC method which consists of the controller, Gc (s ) ,
the delay of PWM, Gd (s ) , and the filter, Gf (s ) in the inverter’s current control 7.4.2. Hysteresis controller
loop (Teodorescu et al., 2011). Typically, the hysteresis controller is implemented in the voltage-
source inverter (VSI) to generate switching pulses and compare the grid
ki s and reference currents as illustrated in Fig. 20(a). Hence, the current
kp + 0 0
s2 + 2 can be controlled within the upper and lower boundaries of the hys-
Gabc (s ) = 0 kp +
ki s
0 teresis band (HB) as depicted in Fig. 20(b). It will control the inverter to
s2 + 2
ki s keep tracking between the output and the reference currents. In other
0 0 kp +
s2 + 2 (43) application, this controller is employed to find out the differences be-
tween the current error and the fixed HB as stated in Eq. (45).
1, ifiref iref > HB

Sn =
7.4.1. Dead-Beat (DB) 1, ifiref iref < HB (45)
The name of “Dead-Beat” is given due to its control operation nul-
ling the error at the end of two sampling periods as shown in Fig. 19. where Sn is the switch signal (n represent phases), iref is the reference
current and iref is the inductor current of n phase.
This controller has the advantage of its high sensitivity to parameter
mismatch and modelling since it is built based on filter and grid model.
7.4.3. Predictive controller (PC)
Furthermore, the DB controller also has a faster transient response
The purpose of this controller is the implementation of a prediction
among the other non-linear controllers when the controller is fine-
model for the future behavior of the variables. This controller applies
tuned. The underlining merit of this controller is its high performance
the same concept in the hysteresis controller, the difference is the PC
in limiting the peak current during grid faults. The implementation of
controls the inverter voltage to drive the grid current to follow the
the DB controller can be made by applying the controller transfer
reference current without having the upper and lower bands as illu-
function in Eq. (44).
strated in Fig. 21. Yu and Chang (2005) have improved the PC algo-
rithm as expressed in Eq. (46). This improvement overcomes the poor
Fig. 17. State feedback control system (Taveiros et al., 2015).
863
Fig. 18. Block diagram for non-linear current controllers with abc -control (Hassaine et al., 2014).
performance under variable parameters in the conventional PC. linearization to achieve an equivalent linear controller. The FLS trans-
form the nonlinear parameters in the dq reference frame to be able to
Iref [n 1] Il [n]
Vo_avg [n] = 1.5Vg [n] 0.5Vg [n 1] + L control the active and reactive currents as depicted in Fig. 23. The main
Tsw (46)
purpose of implementing the FLS is to obtain the dynamic stability
where Vo_avg [n]is the average inverter output voltage, Vg [n] is the despite of the complexity of the controller. This controller is derived
average grid voltage and Iref , Il is the reference current and measured with the expression of the input and output power relation to the in-
current at load, respectively. verter to achieve the linearization phase. Hence, Eqs. (50) and (51) are
expressing the derivation of the nonlinear parameters into the linear
7.4.4. Sliding mode control (SMC) output to achieve the feedback linearization (Soufi et al., 2016; Sekhar
Sliding mode control (SMC) as elaborated in Abadlia et al. (2017), and Mishra, 2014; Montoya et al., 2018; Lalili et al., 2011).
Eddine et al. (2016) and Liu et al. (2020), is widely used in various
x = f (x ) + g (x ). u ; y = h (x ) (50)
fields due to the simplicity and stability controller. The fundamental of
the controller is that the implementation of first order and second order where x is the state vector, u is the input vector, y is the system output,
SMC to cater with the drawbacks of undesirable chattering and dis- f (x ) , h (x ) and g (x ) are vector fields. In order to achieve the output
continuous control function, respectively. The grid current id and iq feedback linearization, the system output should be differentiated to
should be controlled to track the active and reactive reference currents obtain the input parameters.
id and iq , respectively, where id is acquired from the voltage loop and iq m
is manually set to create a desired instantaneous reactive power, q as yi(ri) = Lfri hi + L gi Lfri 1 hi. uj
illustrated in Fig. 22. j=1 (51)
id =
p
; iq =
q where ri is the integer, Lf is the order of Lie derivatives.
vd vd (47)
7.4.6. Backstepping control (BC)
sd = id id ; sq = iq iq (48)
A nonlinear controller of backstepping control (BC) (Kammoun
where sd and sq are sliding mode for active and reactive current, re- et al., 2017; Yazdani and Ferdowsi, 2019) is imposed during the grid
spectively. The sliding mode derivatives are define as: faults to achieve the LVRT operation under control by controlling the
r vd
power flows in the inverter as shown in Fig. 24. The controller is re-
sd id + L i d L
iq VDC ud cognized as a robust and high dynamic controller to cater with severe
= +
sq r
iq + L iq
vq
id L uq uncertainty of parameters. For instance, the controller is capable to
L (49)
limit inrush currents during the grid faults by adjusting the power flows
where ud and uq are the designed controllers for the active and reactive in the inverter. This technique is important to ensure the controller is
current, respectively. efficiently controlling the dynamics of the GCPV inverter in the dq re-
ference frame during and clearing of grid faults, and maintaining the
7.4.5. Feedback linearization strategy (FLS) currents at their levels before the faults occurred. The grid active, Pg
The fundamental of the controller is by imposing the feedback and reactive power, Qg are formulated by the Eqs. (52) and (53).
Fig. 19. (a) Block diagram for DB and (b) the control operation nulling the error after two samples (Teodorescu et al., 2011).
864
Fig. 20. (a) Block diagram for hysteresis controller, (b) the current controller operation waveform (Parvez et al., 2016).
3 3
Pg = (Vg id); Qg = (Vg iq)
2 2 (52)
where Vg is the voltage fault terminal while id, iq are the active and
reactive current grid, respectively. By substituting Eq. (52) into Eq.
(53), a set of the dynamic model in the dq reference frame is expressed
as follows:
1
x1 = (P
kx 1 pv
Vdc )
3 3 2 R
x2 = V V cos( )
2L g 0
V
2L g
P
L g
Qg
3 R
x3 = V V sin(
2L g 0
) + Pg Q
L g (53)
7.4.7. Fuzzy logic control (FLC)

Fig. 22. Current tracking loop of the SMC (Liu et al., 2020).
Fuzzy logic control (FLC) is a very prominent controller that has
flexibility and versatility to work on its own as a pure controller or to
work as a hybrid controller with any conventional controllers. The Wi = µi ( V ); i : rule number (55)
controller has the capability as voltage compensation by controlling the
active and reactive power in point of common coupling (PCC) gener- 4. Defuzzification: The center of gravity method is applied for the
ated by the GCPV systems as depicted in Fig. 25 (Mokryani et al., 2012). defuzzication to dictate the output by calculating the duty cycle.
The FLC consists of four classifications to realize the voltage compen-
z. µi (z ) dz
sation as explained below (Rezaie and Kazemi-Rahbar, 2019; Rashid d=
and Ali, 2017): µi (z ) dz (56)
where µi is the parameter of d expressed in terms as variables in the

1. Fuzzification: The inputs are converted into fuzzy variables based on fuzzy rule table.
the pre-defined functions. The following equation is imposed to
obtain the best system performance by implementing Gaussian
7.4.8. Artificial neural network (ANN)
membership functions.
Complex tasks required a powerful, robust and high-reliability
(x c ) 2 controller to perform the operations, for instance, pattern recognition,
f (x ; , c ) = e 2 2 (54) system identification, classification, speech, vision and control system.
where and c are the width of the bell curve and the center of the peak, Artificial neural network (ANN) has been extensively employed to cope
respectively. to those complex tasks and other engineering problems. ANN is a good
candidate to predict the nonlinear parameters such as the current in
2. Fuzzy rule base: Comprises the information of the system to be used GCPV system. However, the ANN is highly dependent on the database
in fuzzy inference. collected from the field or experiments and not appropriate to be used
3. Fuzzy inference: This is the decision unit that defines the degree of in online monitoring of grid faults. Fig. 26 depicts the ANN controller in
conformity, Wi for each fuzzy rules as expressed in the following detecting the grid fault which comprise of two steps, namely the voltage
equation: indicators generation and the grid fault detection. The grid fault de-
tection composed of three feed-forward layers which are the inputs
Fig. 21. Block diagram for predictive controller (Cortes et al., 2008).
865
Fig. 23. Overview of the FLS in GCPV system (Lalili et al., 2011).
(va (t ), vb (t ), vc (t ) ), the hidden layers (ANN1 – ANN7) and output values

(healthy case or grid fault) (Alberdi et al., 2012; Adouni et al., 2016).
7.4.9. Genetic algorithm (GA)

Genetic algorithm (GA) is a simple and powerful method which
integrate with the artificial survival of the best and fittest chromosome
of genetic operators abstracted from nature to produce a robust tech-
nique to perform an optimization process within a system (Vas, 2001).
Fig. 25. The FLC controlling the active and reactive power (Mokryani et al.,
The GA firstly will evaluate the fitness of the chromosome as expressed
2012).
in Eq. (57), followed by the operations of reproduction, crossover and
mutation as described as follows (Vrionis et al., 2014):
Fig. 24. Active and reactive power control in backstepping control (Yazdani and Ferdowsi, 2019).
866
Fig. 26. ANN for grid fault detection (Adouni et al., 2016).
Reproduction: High fitness string will receive multiple copies in the 8.2. Stability
next generation while lower fitness string receives fewer copies or
none at all. A current controller should be developed for stable operations under
Crossover: Taking and splitting the string into two parts at a ran- every grid fault occurrence. Table 7 shows the PR controller has better
domly generated crossover point and merging with another string at stability among the other two controllers based on the larger range in
the same crossover point. those three steps which are the id, iq and vg . The steps are measured
Mutation: A random alteration of the string which assists in from 0 to 1 pu. The hysteresis controller followed as the second stable
maintaining diversity of the population. Finally, a new population controller despite of its nonlinearity that is prone to distortions. Fig. 27
will be formed to replace the old population. illustrated the current step in which the PR controller produced a more
2 2
stable and damped response under the grid fault condition. Whereas,
Vdc, max Vdc _ss ipv, max ipv _ss the PI controller contribute to a higher overshoot while the hysteresis
fitness = +
Vdc _mv Vdc _ss ipv _mv ipv _ss (57) has a faster response with low oscillation and short overshoot
(Midtsund et al., 2010).
where ss represents the steady state value, mv stands for the maximum
acceptable value, specified by the manufacturer, ipv indicates the pho- 8.3. Robustness
tovoltaic current.
Controller robustness is defined for being strong and in good con-
8. Performance comparison of current controllers dition after the fault occurrence. The CRP, DB, hysteresis, SMC, BC and
GA are among the controllers with strong robustness in current injec-
In this section, a brief overview of performance comparison of tion to cope with the grid fault. Fig. 28 depicts the best example of a
current controllers is presented. The comparison will cover the aspect of robust controller where the unbalance and oscillating power have been
complexity, stability, robustness and power quality assessment. Then, greatly reduced after the BC is imposed into the system.
the summary of the comparison is tabulated in Table 8 which comprises
of merits and demerits of the current controllers. 8.4. Power quality assessment
8.1. Complexity The undesirable chattering issue in the SMC leads to low accuracy
and affected the power quality caused by fast dynamics in sensors and
Many studies with various current controllers have been made data processors. In the case of chattering excites, the harmonic
where the complexity is one of the most important aspects to define the
outcome of the controllers. Controllers like the PNSC, MCC, PR, FLS, Table 7
Stability limit of different controllers in grid inductance (Midtsund et al., 2010).
ANN possess high complexity controllers but they contribute to other
merits such as good stability, high accuracy and efficiency. Contrarily, Method Step in id Step in iq Step in vg
simple controllers are typically chosen based on the application needs
PI controller 0.32 pu 0.29 pu 0.25 pu
where the BPSC, PI, SFC, SMC and BC are identified to have the sim-
PR controller 0.46 pu 0.36 pu 0.36 pu
plicity of the controllers. In spite of the simplicity, the controllers are Hysteresis 0.38 pu 0.30 pu 0.29 pu
still able to carry out the task of stability and efficiency satisfactorily.
867
Table 8
Comparison table of current controllers for current injection techniques.
Current Injection Techniques Current Controllers Merits Demerits
Current compensation PNSC (Saccomando and Svensson, 2001; Teodorescu et al., 2011; • Good stability • High peak current
controls Alepuz, 2009; Chou et al., 2011) • Low power ripples • High complexity controller
• Good elimination in power
oscillation
• Controlling PNS components
PCLC 7(Chen et al., 2016a, 2016b) • Low peak current • Controlling negative-sequence
• Simple controller component only
ARPC (Chen et al., 2016a) • Low power ripples • High complexity controller
• Controlling PNS components
IARC (Teodorescu et al., 2011) • Good harmonic compensation • Oscillation in active power
• High efficiency controller
• Good stability
• Ability to inject active and
reactive current
• Controllability
AARC (Teodorescu et al., 2011) • Good elimination in power
oscillation
• High peak current
BPSC (Teodorescu et al., 2011) • Simple controller • High peak current
• Good stability • Controlling
component only
positive-sequence
RCI CAP (Jia et al., 2018; Brandao et al., 2019; Oon et al., 2018) • Good elimination in power • High cost
oscillation • Prone to trigger overcurrent limit
• High energy yield • Risk of overheat to maintain the
constant output power
CRP (Jia et al., 2018; Oon et al., 2018) • Simple reactive current control • Cannot generate active power
• Robustness • Limited to constant power only
CAC (Yang et al., 2014a; Brandao et al., 2019; Oon et al., 2018) • Low power operation • Prone
limit
to trigger overcurrent
• Limited to constant power only

• Require power derating operation
• Limited to small voltage range
VAC (Oon et al., 2018) • Good overcurrent protection • Fault in low current
• Simple controller • Low accuracy
• High peak current
CPC (Yang et al., 2014a) • Good stability • High peak current
• Good overcurrent protection
• Applicable for all inverter
MCC (Oon et al., 2018) • High accuracy • High complexity controller
• Good overcurrent protection • Highly rely on inverter design
• Limited
current
selection of maximum
Linear current controls PI-based (Parvez et al., 2016; Teodorescu et al., 2011; Hassaine • Simple controller • Poor elimination in steady-state
et al., 2014; Blaabjerg et al., 2006; Timbus et al., 2009) • Easy to control error
• Good filtering • Poor harmonic compensation
PR-based (Parvez et al., 2016; Teodorescu et al., 2011; Hassaine
et al., 2014; Blaabjerg et al., 2006)
• High gain in resonance
frequency
• Hardware complexity
• High dynamic response
• Good
error
elimination in steady-state
• Very good harmonic

compensation
RC (Parvez et al., 2016; Teodorescu et al., 2011; Gonçalves et al., • Good elimination in steady-state • Slow dynamic response
2016; Timbus et al., 2009) error • Low stability
• High gain in resonance
frequency
• Easy reference current tracking
• Good harmonic compensation
SFC (Yang et al., 2017; Taveiros et al., 2015) • Simplicity • Less sensitive to disturbance
• Robustness • Less accurate & efficiency
• Low computational cost
Non-linear current controls DB (Parvez et al., 2016; Teodorescu et al., 2011; Cortes et al., • High sensitivity • High delay
2008; Hassaine et al., 2014; Blaabjerg et al., 2006; Timbus et al., • Good transient response • Limited to active filter application
2009) • Robustness only
• High performance
• High dynamic and fast response
Hysteresis (Parvez et al., 2016; Prasad et al., 2016; Cortes et al., • Robustness • Varying modulation frequency
2008; Dai et al., 2009; Meo and Perfetto, 2002; Hassaine et al., • Good stability • High complexity controller
2014; Blaabjerg et al., 2006) • High dynamic • Unwanted resonance in the grid
• Good transient response voltage
• Individual load parameters
PC (Parvez et al., 2016; Meo and Perfetto, 2002; Cortes et al., • Minimize forecast error • Poor performance under variable
2008; Yu and Chang, 2005) • Good precision control parameters
• Low harmonic & noise • Hardware complexity
• Good dynamic response • High sampling rate
(continued on next page)
868
Table 8 (continued)
Current Injection Techniques Current Controllers Merits Demerits
SMC (Abadlia et al., 2017; Liu et al., 2020) • Simple controller • Chattering
• Robustness • Discontinuous control function
• Good stability
• High speed controller
FLS (Soufi et al., 2016; Sekhar and Mishra, 2014; Montoya et al., • Good stability • Complex controller
2018; Lalili et al., 2011) • High accuracy • High computational process
• Good dynamic response
BC (Kammoun et al., 2017; Yazdani and Ferdowsi, 2019) • Robustness • Adjustment of gains
• High stability
• Simple controller
• High efficiency
FLC (Mokryani et al., 2012; Rezaie and Kazemi-Rahbar, 2019) • Comprehensive controller • Complex controller
• High efficiency • Dependent on the fuzzy rules
• Easy integration with
conventional controllers
• Noisy-friendly controller
• Low overshoot & oscillation
• Fast convergence
• Parameter insensitive
ANN (Alberdi et al., 2012; Adouni et al., 2016) • Powerful controller • Complex controller
• Used for complex problems • High computational time
• Parallelism & high reliability
GA (Vas, 2001; Vrionis et al., 2014) • Vast solution set • Complicated method
• Good optimization method • High computational time
• Used for complex problems
Fig. 27. Current step response in the PLL connected to grid faults (Midtsund et al., 2010).
distortion will be elevated. State feedback and PI controllers presented reviewed. A detailed discussion of previous techniques has also been
low harmonic distortion as delineated in Fig. 29. The figure is clearly discussed. Finally, several research challenges and possible future re-
shown the SMC is badly affected in harmonic due to the chattering issue search directions on the LVRT have been proposed. The possible chal-
and heavily impacting the power quality. lenges to implementing the current controllers with a proper selection
of inverter topologies are highlighted as follows:
9. Key challenges and future recommendations
• The main issue of the grid fault is caused by the unbalance grid due
to natural interference, exposure to technical errors or connection of
In the previous section, the existing work on the LVRT has been
869
Fig. 28. Robustness of backstepping control in unbalance and power compensation, (a) oscillations in active and reactive power without BC, and (b) smooth and
robust output with BC (Yazdani and Ferdowsi, 2019).
the power quality as normal operation. Many current controllers are

trying to inject symmetrical and sinusoidal current to support the
grid faults, but some of the controllers ignore the reactive power
support behavior.
• The selection of the current controllers are defined on several factors
such as the performance (complexity, stability, robustness and
power quality), system architecture and cost. The IARC is reported
to have more benefits than the other current compensation controls.
The CPC from reactive current injection technique is chosen to due
to the stability possess by this controller. The linear current controls
presented four controllers and the PR-based is the most outstanding
among the others. There are nine controllers have been discussed in
non-linear current controls and several controllers show great im-
Fig. 29. Measurement of total harmonic distortion (THD) of three controllers; pacts like the DB, BC and FLC controllers. But still, all the controllers
SF: state feedback, PI: proportional integral and SM: sliding mode (Taveiros own the drawbacks that will be affecting the strategy and equip-
et al., 2015). ment. Therefore, a study can be initiated for hybrid controllers to
complement the drawbacks owns by the current controllers.
a large number of loads to the power grid. It will cause the inter- • There are few reports and studies on the implementation of the FLC,
ruption in current flows of the inverter within a few milliseconds ANN and GA for the LVRT in the GCPV systems. The wind power
and trigger either the LVRT, short-term islanding or the anti-is- generation is more advanced in utilizing these controllers.
landing itself. This event occurs instantly, hence, the selection of Therefore, they should be given more attention in future studies to
inverter topology plays an important role in the current controllers enhance the existing current controllers.
to inject a sufficient amount of active and reactive current to cope
with the grid faults. The researchers should identify the benefits and
10. Conclusion
drawbacks of the topology and current controller to integrate with,
for instance, a hybrid inverter topology of soft-switching bidirec-
This paper presents a comprehensive review on various current in-
tional inverter employed the high-frequency switching (from
jection techniques and controllers associated with the LVRT control
100 kHz to 10 MHz) which can cause a surge of voltage and current
strategy in the GCPV systems. The main objective is to discuss the
during the switching events. Thus, a robust topology and protective
capabilities of the controllers to deliver the required active or reactive
mechanism are required to prevent severe damage to the equipment.
•
current compensations to support the power grid during the exposure to
The LVRT is a safety feature introduced in the last decade to im-
grid faults. Based on the discussion, a comparative analysis has been
provise the existing anti-islanding feature. The great outcome is, the
made to highlight the merits and demerits of each controller, while
GCPV system is not necessary to disconnect for every fault occur-
considering the performance parameters of reliability, robustness, sta-
rence where the inverter protection system will automatically
bility, dynamic response, and harmonic compensation capability.
identify the faults based on the grid codes requirement before any
Although the PNSC, IARC, PR, DB, hysteresis, BC and FLC controllers
disconnection occurs. The recent grid codes have included the grid
offer a credible performance, implementation complexity remains a
owner and GCPV power plants to comply with the THD and remain
noticeable demerit. In terms of less complexity, stability, robustness
870
and power quality, the BC and FLC controllers, relatively, presents Chatterjee, A., Mohanty, K.B., 2016. Control of single phase power inverter using model
improved control strategies against other current compensation tech- predictive controller for grid integrated renewable energy systems. IEEE Trans.
Chen, H.-C., Lee, C.-T., Cheng, P.-T., Teodorescu, R., Blaabjerg, F., 2016a. A low-voltage
niques under consideration. ride-through technique for grid-connected converters with reduced power transistors
stress. IEEE Trans. Power Electron., pp. 1–1, doi: 10.1109/tpel.2016.2522511.
Declaration of Competing Interest Chen, H.-C., Wu, P.-H., Cheng, P.-T., 2016b. A transformer inrush reduction technique for
low-voltage ride-through operation of renewable converters. IEEE Trans. Ind. Appl.
52 (3), 2467–2474. https://doi.org/10.1109/tia.2016.2533497.
The authors declare that they have no known competing financial Cho, Y.-W., Cha, W.-J., Kwon, J.-M., Kwon, B.-H., 2016. High-efficiency bidirectional
DAB inverter using a novel hybrid modulation for stand-alone power generating
interests or personal relationships that could have appeared to influ- system with low input voltage. IEEE Trans. Power Electron. 31 (6), 4138–4147.
ence the work reported in this paper. https://doi.org/10.1109/tpel.2015.2476336.
Chou, S.-F., Lee, C.-T., Cheng, P.-T., Blaabjerg, F., 2011. A reactive current injection
technique for renewable energy converters in low voltage ride-through operations.
Acknowledgements
IEEE Trans..
Cortes, P., Kazmierkowski, M.P., Kennel, R.M., Quevedo, D.E., Rodriguez, J., 2008.
The authors would like to thank the technical and financial assis- Predictive control in power electronics and drives. IEEE Trans. Ind. Electron. 55 (12),
tance of UM Power Energy Dedicated Advanced Centre (UMPEDAC) 4312–4324. https://doi.org/10.1109/tie.2008.2007480.
Dai, W., Wang, B., Yang, H., 2009. A hysteretic current controller for active power filter
and the Higher Institution Centre of Excellence (HICoE) Program with constant frequency. In: International Conference on Computational Intelligence
Research Grant, UMPEDAC - 2018 (MOHE HICOE - UMPEDAC), for Measurement Systems and Applications.
Ministry of Education Malaysia, TOP100UMPEDAC, RU012-2019 and Dhimish, M., Holmes, V., Mehrdadi, B., Dales, M., 2017a. Multi-layer photovoltaic fault
detection algorithm. High Voltage 2 (4), 244–252. https://doi.org/10.1049/hve.
the Fundamental Research Grant Scheme (FRGS), FP103-2018A, 2017.0044.
University of Malaya. Dhimish, M., Holmes, V., Mehrdadi, B., Dales, M., 2017b. Simultaneous fault detection
algorithm for grid-connected photovoltaic plants. IET Renew. Power Gener. 11 (12),
1565–1575. https://doi.org/10.1049/iet-rpg.2017.0129.
References Dogga, R., Pathak, M., 2019. Recent trends in solar PV inverter topologies. Sol. Energy
183, 57–73. https://doi.org/10.1016/j.solener.2019.02.065.
Abadlia, I., Adjabi, M., Bouzeria, H., 2017. Sliding mode based power control of grid- E. Commission, 2016. Grid Code for Peninsular Malaysia.
connected photovoltaic-hydrogen hybrid system. Int. J. Hydrogen Energy 42 (47), E. Commission, 2017. Grid Code for Sabah and Labuan.
28172–28182. https://doi.org/10.1016/j.ijhydene.2017.08.215. Eddine, K.D., Mezouar, A., Boumediene, L., Bossche, A.P.M.V.D., 2016. A comprehensive
Adouni, M., Chariag, D., Diallo, D., Ben Hamed, M., Sbita, L., 2016. FDI based on artificial review of LVRT capability and sliding mode control of grid-connected wind-turbine-
neural network for low-voltage-ride-through in DFIG-based wind turbine. ISA Trans. drive doubly fed induction generator. Taylor and Francis Group LLC.
64, 353–364. https://doi.org/10.1016/j.isatra.2016.05.009. Elyaalaoui, K., Ouassaid, M., Cherkaoui, M., 2019. Dispatching and control of active and
Ahmad, Z., Singh, S., 2017. Comparative analysis of single phase transformerless inverter reactive power for a wind Farm considering fault ride-through with a proposed PI
topologies for grid connected PV system. Sol. Energy 149, 245–271. https://doi.org/ reactive power control. Renew. Energy Focus 28, 56–65. https://doi.org/10.1016/j.
10.1016/j.solener.2017.03.080. ref.2018.11.002.
Akagi, H., Watanabe, E.H., Aredes, M., 2007. Instantaneous Power Theory and Farhang, J., Eydi, M., Asaei, B., Farhangi, B., 2015. Flexible strategy for active and re-
Applications to Power Conditioning. Wiley & Sons Inc. Publication. active power control in grid connected inverter under unbalanced grid fault. In: 2015
Alberdi, M., Amundarain, M., Garrido, A., Garrido, I., 2012. Neural control for voltage 23rd Iranian Conference on Electrical Engineering (ICEE).
dips ride-through of oscillating water column-based wave energy converter equipped Fu, R., Feldman, D., Margolis, R., 2018. U.S. Solar Photovoltaic System Cost Benchmark:
with doubly-fed induction generator. Renew. Energy 48, 16–26. https://doi.org/10. Q1 2018.
1016/j.renene.2012.04.014. Galvez-Carrillo, M., Keyser, R.D., Ionescu, C., 2009. Nonlinear predictive control with
Alepuz, S., et al., 2009. Control strategies based on symmetrical components for grid- dead-time compensator: application to a solar power plant. Sol. Energy 83 (5),
connected converters under voltage dips. IEEE Trans. Ind. Electron. 56 (6), 743–752.
2162–2173. https://doi.org/10.1109/tie.2009.2017102. Garnica Lopez, M.A., Garcia de Vicuna, J.L., Miret, J., Castilla, M., Guzman, R., 2018.
Almeida, P.M., Monteiro, K.M., Barbosa, P.G., Duarte, J.L., Ribeiro, P.F., 2016. Control strategy for grid-connected three-phase inverters during voltage sags to meet
Improvement of PV grid-tied inverters operation under asymmetrical fault conditions grid codes and to maximize power delivery capability. IEEE Trans. Power Electron.
_ Elsevier Enhanced Reader. Sol. Energy. 33 (11), 9360–9374. https://doi.org/10.1109/tpel.2018.2792478.
Al-Shetwi, A.Q., Sujod, M.Z., Blaabjerg, F., 2018. Low voltage ride-through capability Gonçalves, A.F.Q., Aguiar, C.R., Bastos, R.F., Pozzebon, G.G., Machado, R.Q., 2016.
control for single-stage inverter-based grid-connected photovoltaic power plant. Sol. Voltage and power control used to stabilise the distributed generation system for
Energy 159, 665–681. https://doi.org/10.1016/j.solener.2017.11.027. stand-alone or grid-connected operation. IET Power Electron. 9 (3), 491–501.
Al-Shetwi, A.Q., Sujod, M.Z., Blaabjerg, F., Yang, Y., 2019. Fault ride-through control of https://doi.org/10.1049/iet-pel.2015.0071.
grid-connected photovoltaic power plants: a review. Sol. Energy 180, 340–350. Hassaine, L., Olias, E., Quintero, J., Salas, V., 2014. Overview of power inverter topolo-
https://doi.org/10.1016/j.solener.2019.01.032. gies and control structures for grid connected photovoltaic systems. Renew. Sustain.
Ammiche, M., Kouadri, A., Halabi, L.M., Guichi, A., Mekhilef, S., 2018. Fault detection in Energy Rev. 30, 796–807. https://doi.org/10.1016/j.rser.2013.11.005.
a grid-connected photovoltaic system using adaptive thresholding method. Sol. Huka, G.B., Li, W., Chao, P., Peng, S., 2018. A comprehensive LVRT strategy of two-stage
Energy 174, 762–769. https://doi.org/10.1016/j.solener.2018.09.024. photovoltaic systems under balanced and unbalanced faults. Electric Power Energy
Babaie, E., Asl, E., 2016. A new topology for Z-source half-bridge inverter with low Syst.
voltage stress on capacitors. Electr. Power Syst. Res. 140, 722–734. https://doi.org/ Hunter, G., Andrade, I., Riedemann, J., Blasco-Gimenez, R., Peña, R., 2016. Active and
10.1016/j.epsr.2016.04.010. reactive power control during unbalanced grid voltage in PV systems. In: IECON Proc.
Blaabjerg, F., Teodorescu, R., Liserre, M., Timbus, A.V., 2006. Overview of control and (Industrial Electron. Conf.), pp. 3012–3017.
grid synchronization for distributed power generation systems. IEEE Trans. Ind. Hyo-Sang, C., Yong-Sun, C., Byung-Ik, J., Dong-Chul, C., Fathy, A., 2010. Comparison of
Electron. 53 (5), 1398–1409. https://doi.org/10.1109/tie.2006.881997. the unbalanced faults in three-phase resistive and matrix-type SFCLs. IEEE Trans.
Bode, G.H., Loh, P.C., Newman, M.J., Holmes, D.G., 2005. An improved robust predictive Appl. Supercond. 20 (3), 1215–1218. https://doi.org/10.1109/tasc.2010.2043662.
current regulation algorithm. IEEE Trans. Ind. Appl. 41 (6), 1720–1733. https://doi. Islam, F.M.R., Mamun, K.A., 2017. Smart Energy Grid Design for Island Countries.
org/10.1109/tia.2005.858324. Springer Nature.
Brandao, D.I., Mendes, F.E.G., Ferreira, R.V., Silva, S.M., Pires, I.A., 2019. Active and Islam, M., Mekhilef, S., Hasan, M., 2015. Single phase transformerless inverter topologies
reactive power injection strategies for three-phase four-wire inverters during sym- for grid-tied photovoltaic system - a review. Renew. Sustain. Energy Rev. 45, 69–86.
metrical/asymmetrical voltage sags. IEEE Trans. Ind. Appl. 55 (3), 2347–2355. https://doi.org/10.1016/j.rser.2015.01.009.
https://doi.org/10.1109/tia.2019.2893135. Jana, J., Saha, H., Das Bhattacharya, K., 2017. A review of inverter topologies for single-
Byung-Ik, J., Hyo-Sang, C., Yong-Sun, C., Dong-Chul, C., 2012. Analysis of the unbalanced phase grid-connected photovoltaic systems. Renew. Sustain. Energy Rev. 72,
fault in three-phase flux-coupling type SFCL using the symmetrical coordinate 1256–1270. https://doi.org/10.1016/j.rser.2016.10.049.
method. IEEE Trans. Appl. Supercond. 22 (3), 5601305. https://doi.org/10.1109/ Jerin, R.A., Thomas, M., Kaliannan, P., Subramaniam, U., 2018. Enhancing low voltage
tasc.2012.2184510. ride through capability in utility grid connected single phase solar photovoltaic
Cabrera-Tobar, A., Bullich-Massague, E., Aragues-Penalba, M., Gomis-Bellmunt, O., 2016. system. J. Eng. Sci. Technol. 13.
Review of advanced grid requirements for the integration of large scale photovoltaic Jia, J., Yang, G., Nielsen, A.H., 2018. A review on grid-connected converter control for
power plants in the transmission system. Renew. Sustain. Energy Rev. short-circuit power provision under grid unbalanced faults. IEEE Trans. Power
Carrasco, J.E.-G., Tena, J.M., Ugena, D., Alonso-Martinez, J., Santos-Martin, D., Arnaltes, Delivery 33 (2), 649–661. https://doi.org/10.1109/tpwrd.2017.2682164.
S., 2013. Testing low voltage ride through capabilities of solar inverters. Electr. Kala, P., Arora, S., 2017. A comprehensive study of classical and hybrid multilevel in-
Power Syst. Res. 96, 111–118. https://doi.org/10.1016/j.epsr.2012.10.011. verter topologies for renewable energy applications. Renew. Sustain. Energy Rev. 76,
Chakraborty, S., Hassan, M., Razzak, M., 2017. Transformer-less single-phase grid-tie 905–931. https://doi.org/10.1016/j.rser.2017.02.008.
photovoltaic inverter topologies for residential application with various filter circuits. Kammoun, S., Sallem, S., Kammoun, M., 2017. Backstepping control for low-voltage ride
Renew. Sustain. Energy Rev. 72, 1152–1166. https://doi.org/10.1016/j.rser.2016. through enhancement of DFIG-based wind turbines. Arab. J. Sci. Eng. 42 (12),
10.032. 5083–5099.
871
Karanayil, B., Pou, J., Mirhosseini, M., Agelidis, V.G., 2014. Low-voltage ride through Ramirez, D., Martinez-Rodrigo, F., de Pablo, S., Carlos Herrero-de Lucas, L., 2017.
capability of three-phase grid-connected photovoltaic inverters with slim film capa- Assessment of a non linear current control technique applied to MMC-HVDC during
citors. In: 2014 IEEE Energy Convers. Congr. Expo. ECCE, pp. 32–38. grid disturbances. Renew. Energy 101, 945–963. https://doi.org/10.1016/j.renene.
Krishna, R., Kottayil, S.K., Leijon, M., 2010. Predictive current controller for a grid 2016.09.050.
connected three level inverter with reactive power control. In: Presented at the 2010 Rashid, G., Ali, M., 2017. Fault ride through capability improvement of DFIG based wind
IEEE 12th Work. Control Mode. Power Electron, COMPEL. farm by fuzzy logic controlled parallel resonance fault current limiter. Electr. Power
Lalili, D., Mellit, A., Lourci, N., Medjahed, B., Berkouk, E., 2011. Input output feedback Syst. Res. 146, 1–8. https://doi.org/10.1016/j.epsr.2017.01.018.
linearization control and variable step size MPPT algorithm of a grid-connected Razali, A.M., Rahman, M.A., Rahim, N.A, 2014. Implementation of d-q decoupling and
photovoltaic inverter. Renew. Energy 36 (12), 3282–3291. https://doi.org/10.1016/ feed-forward current controller for grid connected three phase voltage source con-
j.renene.2011.04.027. verter. In: IECON Proc. (Industrial Electron. Conf.), pp. 1733–1739.
Lee, C.-T., Hsu, C.-W., Cheng, P.-T., 2011. A low-voltage ride-through technique for grid- Rezaie, H., Kazemi-Rahbar, M., 2019. Enhancing voltage stability and LVRT capability of
connected converters of distributed energy resources. IEEE Trans. Ind. Appl. 47 (4), a wind-integrated power system using a fuzzy-based SVC. Eng. Sci. Technol. 22 (3),
1821–1832. https://doi.org/10.1109/tia.2011.2155016. 827–839.
Lin, X., Han, Y., Yang, P., Wang, C., Xiong, J., 2018. Low-voltage ride-through techniques Roberts, M.B., Bruce, A., MacGill, I., 2019. Impact of shared battery energy storage sys-
for two-stage photovoltaic system under unbalanced grid voltage sag conditions. In: tems on photovoltaic self-consumption and electricity bills in apartment buildings.
2018 IEEE 4th South. Power Electron. Conf. SPEC 2018, 2019, 2018. Appl. Energy 245, 78–95. https://doi.org/10.1016/j.apenergy.2019.04.001.
Liu, J., Luo, W., Gao, Y., Yin, Y., Sun, G., 2020. Sliding mode control of grid-connected RodrÍguez, P., Timbus, A., Teodorescu, R., Liserre, M., Blaabjerg, F., 2006. Independent
power converters for microgrid applications. In: Meng, W., Wang, X., Liu, S. (Eds.), PQ control for distributed power generation systems under grid faults. In: IECON
Distributed Control Methods and Cyber Security Issues in Microgrids. Academic Press Proc. (Industrial Electron. Conf.), no. 2, pp. 5185–5190.
(Chapter 1). RodrÍguez, P., Timbus, A., Teodorescu, R., Liserre, M., Blaabjerg, F., 2009. Reactive
Liu, X., 2013. Power control of single-stage PV inverter for distribution system. Master of power control for improving wind turbine system behavior under grid faults. IEEE
Science in Electrical Engineering, Electrical and Computer Engineering, University of Trans. Power Electron. 24 (7), 1798–1801. https://doi.org/10.1109/tpel.2009.
Kentucky. 2014650.
Luo, X., et al., 2018. Review of voltage and frequency grid code specifications for elec- Saccomando, G., Svensson, J., 2001. Transient operation of grid-connected voltage source
trical energy storage applications. Energies 11 (5), pp. https://doi.org/10.3390/ converter under unbalanced voltage conditions. IEEE Trans.
en11051070. Sadeghkhani, I., Hamedani Golshan, M.E., Mehrizi-Sani, A., Guerrero, J.M., 2018. Low-
Magueed, F.A., Sannino, A., Svensson, J., 2004. Transient performance of voltage source voltage ride-through of a droop-based three-phase four-wire grid-connected micro-
converter under unbalanced voltage dips, pp. 1163–1168. grid. IET Gener. Transm. Distrib. 12 (8), 1906–1914. https://doi.org/10.1049/iet-
Malesani, L., Mattavelli, P., Buso, S., 1999. Robust dead-beat current control for PWM gtd.2017.1306.
rectifiers and active filters. IEEE Trans. Ind. Appl. 35 (3), 613–620. Sajedi, S., Farrell, M., Basu, M., 2019. DC side and AC side cascaded multilevel inverter
Mattavelli, P., Spiazzi, G., Tenti, P., 2003. Predictive digital control of power factor topologies - a comparative study due to variation in design features. Int. J. Electr.
preregulators using disturbance observer for input voltage estimation. Power Energy Syst. 113, 56–70. https://doi.org/10.1016/j.ijepes.2019.05.019.
Mattavelli, P., Spiazzi, G., Tenti, P., 2005. Predictive digital control of power factor Sekhar, P., Mishra, S., 2014. Sliding mode based feedback linearizing controller for grid
preregulators with input voltage estimation using disturbance observers. IEEE Trans. connected multiple fuel cells scenario. Int. J. Electr. Power Energy Syst. 60, 190–202.
Power Electron. 20 (1), 140–147. https://doi.org/10.1109/tpel.2004.839821. https://doi.org/10.1016/j.ijepes.2014.02.007.
Meo, S., Perfetto, A., 2002. Comparison of different control techniques for active filter Seung-Tak, K., Byung-Kwan, K., Sun-Ho, B., Jung-Wook, P., 2013. Application of SMES
applications. In: ICCDCS 2002 - 4th IEEE Int. Caracas Conf. Devices, Circuit Syst., and grid code compliance to wind/photovoltaic generation system. IEEE Trans. Appl.
Caracas, vol. 45, pp. 722–729. Supercond. 23 (3), 5000804. https://doi.org/10.1109/tasc.2012.2232962.
Meral, M., Celik, D., 2019. A comprehensive survey on control strategies of distributed Shin, D., Lee, K.-J., Lee, J.-P., Yoo, D.-W., Kim, H.-J., 2015. Implementation of fault ride-
generation power system under normal and abnormal conditions. Annu. Rev. through techniques of grid-connected inverter for distributed energy resources with
Control 15. adaptive low-pass notch PLL. IEEE Trans. Power Electron. 30 (5), 2859–2871.
Midtsund, T., Suul, J., Undeland, T., 2010. Evaluation of current controller performance https://doi.org/10.1109/tpel.2014.2378792.
and stability for voltage source converters connected to a weak grid. In: 2nd IEEE Shukla, A., Ghosh, A., Joshi, A., 2008. Improved multilevel hysteresis current regulation
International Symposium on Power Electronics for Distributed Generation, HeFei, and capacitor voltage balancing schemes for flying capacitor multilevel inverter. IEEE
China, 2010. IEEE, pp. 382–388, doi: 10.1109/PEDG.2010.5545794. Trans. Power Electron. 23 (2), 518–529. https://doi.org/10.1109/tpel.2007.915788.
Miret, J., Camacho, A., Vicuña, L.G.d., Matas, J., 2013. Control scheme with voltage Sosa, J.L., Castilla, M., Miret, J., Matas, J., Al-Turki, Y.A., 2016. Control strategy to
support capability for distributed generation inverters under voltage sags. IEEE maximize the power capability of PV three-phase inverters during voltage sags. IEEE
Trans. Trans. Power Electron. 31 (4), 3314–3323. https://doi.org/10.1109/tpel.2015.
Miret, J., García de Vicuña, J.L., de la Hoz, J., Camacho, A., Castilla, M., 2015. Reactive 2451674.
current injection protocol for low-power rating distributed generation sources under Soufi, Y., Kahla, S., Bechouat, M., 2016. Feedback linearization control based PSO for
voltage sags. IET Power Electron. 8 (6), 879–886. https://doi.org/10.1049/iet-pel. MPPT of wind turbine equipped by PMSG connected to the grid. Int. J. Hydrogen
2014.0593. Energy 41 (45), 20950–20955. https://doi.org/10.1016/j.ijhydene.2016.06.010.
Mokryani, G., Siano, P., Piccolo, A., Calderaro, V., 2012. A fuzzy logic controller to in- Taveiros, F., Barros, L., Costa, F., 2015. Back-to-back converter state-feedback control of
crease fault ride-through capability of variable speed wind turbines. In: Qing, A. (Ed. DFIG (doubly-fed induction generator)-based wind turbines. Energy 89, 896–906.
), Applied Computational Intelligence and Soft Computing, vol. 2012. Hindawi https://doi.org/10.1016/j.energy.2015.06.027.
Publishing Corporation, p. 10. Teodorescu, R., Liserre, M., Rodríguez, P., 2011. Grid Converters for Photovoltaic and
Montoya, O., Garces, A., Serra, F., 2018. DERs integration in microgrids using VSCs via Wind Power Systems. John Wiley & Sons Ltd.
proportional feedback linearization control - Supercapacitors and distributed gen- Tian, H., Gao, F., Ma, C., He, G., Li, G., 2014. A review of low voltage ride-through
erators. J. Storage Mater. 16, 250–258. https://doi.org/10.1016/j.est.2018.01.014. techniques for photovoltaic generation systems. In: 2014 IEEE Energy Conversion
Nanou, S.I., Papathanassiou, S.A., 2014. Modeling of a PV system with grid code com- Congress and Exposition (ECCE), Pittsburgh, PA. IEEE, pp. 1566–1572.
patibility. Electr. Power Syst. Res. 116, 301–310. https://doi.org/10.1016/j.epsr. Timbus, A., Liserre, M., Teodorescu, R., Rodriguez, P., Blaabjerg, F., 2009. Evaluation of
2014.06.021. current controllers for distributed power generation systems. IEEE Trans. Power
Neumann, T., Erlich, I., 2012. Modelling and control of photovoltaic inverter systems Electron. 24 (3), 654–664. https://doi.org/10.1109/tpel.2009.2012527.
with respect to German grid code requirements. IEEE Trans. Vas, P., 2001. Artificial-Intelligence-Based Drives. In: Rashid, M.H., Irwin, J.D. (Eds.),
Obeidat, F., 2018. A comprehensive review of future photovoltaic systems. Sol. Energy. Power Electronics Handbook. Academic Press, pp. 769–777 (Chapter 29).
Obi, M., Bass, R., 2016. Trends and challenges of grid-connected photovoltaic systems – a Vrionis, T., Koutiva, X., Vovos, N., 2014. A genetic algorithm-based low voltage ride-
review. Renew. Sustain. Energy Rev. 58, 1082–1094. through control strategy for grid connected doubly fed induction wind generators.
Oon, K.H., Tan, C.K., Bakar, A.H.A., Che, H.S., Mokhlis, H., Illias, H.A., 2018. IEEE Trans. Power Syst. 29 (3), 1325–1334. https://doi.org/10.1109/TPWRS.2013.
Establishment of fault current characteristics for solar photovoltaic generator con- 2290622.
sidering low voltage ride through and reactive current injection requirement. Renew. Wang, Y., Song, H., Xu, D., 2019. Soft-switching bidirectional DC/DC converter with an
Sustain. Energy Rev. LCLC resonant circuit. IEEE J. Emerg. Selected Topics Power Electron. 7 (2),
Papanikolaou, N.P., 2013. Low-voltage ride-through concept in flyback inverter-based 851–864. https://doi.org/10.1109/jestpe.2019.2904535.
alternating current–photovoltaic modules. IET Power Electron. 6 (7), 1436–1448. Weise, B., 2015. Impact of K-factor and active current reduction during fault-ride-through
https://doi.org/10.1049/iet-pel.2012.0726. of generating units connected via voltage-sourced converters on power system sta-
Parvez, M., Elias, M.F.M., Rahim, N.A., Osman, N., 2016. Current control techniques for bility. IET Renew. Power Gener. 9 (1), 25–36. https://doi.org/10.1049/iet-rpg.2014.
three-phase grid interconnection of renewable power generation systems: a review. 0116.
Sol. Energy 135, 29–42. https://doi.org/10.1016/j.solener.2016.05.029. Yang, Y., Zhou, K., Blaabjerg, F., 2013. Harmonics suppression for single-phase grid-
Patel, P.V., 2018. Modeling and control of three-phase grid-connected PV inverters in the connected PV systems in different operation modes. Mater. Today Proc. 5,
presence of grid faults. Master of Science in Electrical Engineering, Electrical 1537–1544.
Engineering, Missouri University of Science and Technology. Yang, Y., Wang, H., Blaabjerg, F., 2014a. Reactive power injection strategies for single-
Prabaharan, N., Palanisamy, K., 2017. A comprehensive review on reduced switch mul- phase photovoltaic systems considering grid requirements. IEEE Trans. Ind. Appl. 50
tilevel inverter topologies, modulation techniques and applications. Renew. Sustain. (6), 4065–4076. https://doi.org/10.1109/tia.2014.2346692.
Energy Rev. 76, 1248–1282. https://doi.org/10.1016/j.rser.2017.03.121. Yang, Y., Blaabjerg, F., Wang, H., 2014b. Low-voltage ride-through of single-phase
Prasad, V., Jayasree, P.R., Sruthy, V., 2016. Active power sharing and reactive power transformerless photovoltaic inverters. IEEE Trans. Ind. Appl. 50 (3), 1942–1952.
compensation in a grid-tied photovoltaic system. In: International Conference on https://doi.org/10.1109/tia.2013.2282966.
Processing of Materials, Minerals and Energy, vol. 55, pp. 1537–1544. Yang, B., Hu, Y., Huang, H., Shu, H., 2017. Perturbation estimation based robust state
872
feedback control for grid connected DFIG wind energy conversion system. Int. J. Spec. Conf., 2005, vol. 2005, pp. 231–236.
Hydrogen Energy 42 (33), 20994–21005. https://doi.org/10.1016/j.ijhydene.2017. Zeb, K., et al., 2018. A comprehensive review on inverter topologies and control strategies
06.222. for grid connected photovoltaic system. Renew. Sustain. Energy Rev. 94, 1120–1141.
Yazdani, S., Ferdowsi, M., 2019. Robust backstepping control of synchronverters under https://doi.org/10.1016/j.rser.2018.06.053.
unbalanced grid condition. In: Presented at the IEEE Power and Energy Conference at Zhou, H., Xiao, S., Yang, G., Geng, H., 2012. Modeling and control for a bidirectional
Illinois (PECI), Champaign, IL. buck-boost cascade inverter. IEEE Trans. Power Electron. 27 (3), 1401–1413. https://
Yu, B., Chang, L., 2005. Improved Predictive Current Controlled PWM for Single-Phase doi.org/10.1109/tpel.2010.2103957.
Grid-Connected Voltage Source Inverters. In: PESC Rec. - IEEE Annu. Power Electron.
873

10 1016@j Solener 2020 06 085

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

10 1016@j Solener 2020 06 085

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

10 1016@j Solener 2020 06 085

Uploaded by

Copyright:

Available Formats

Solar Energy 207 (2020) 851–873

Contents lists available at ScienceDirect

A review on current injection techniques for low-voltage ride-through and T

ARTICLE INFO ABSTRACT

Nomenclature AARC Average active reactive control

(Hassan et al. this paper) 2020 • PNSC • PR control 25 Yes

Germany BDEW Power plants connected to the medium-voltage grid 2018

power processing stage, transformer and transformerless inverter,

6.1. Power processing stage

This inverter topology consists of two main types, namely as single-

The transformerless inverter is the most acknowledged inverter in

Fig. 2. Classification of inverter topologies.

p = v· i (1) current, Eq. (1) is further elaborated as follows:

where p is the instantaneous active power, v is the voltage vector in the

Current compensation controls Sequences control of voltage and current

Several current controllers have been proposed to counter the main

7.2.1. Constant active power (CAP)

where Id is defined as the active current, Kp is the active power ratio, Vg

7.2.2. Constant reactive power (CRP)

7.3.1. Proportional-Integral (PI)-based control

a. Addition of the PNS components without voltage feed-forward

7.3.2. Proportional-Resonant (PR)-based control

7.3.3. Resonant control (RC)

The synchronous controller, Gdq (s ) implements the inverse trans-

7.3.4. State feedback control (SFC)

1, ifiref iref > HB

Fig. 17. State feedback control system (Taveiros et al., 2015).

7.4.7. Fuzzy logic control (FLC)

where µi is the parameter of d expressed in terms as variables in the

(va (t ), vb (t ), vc (t ) ), the hidden layers (ANN1 – ANN7) and output values

7.4.9. Genetic algorithm (GA)

• Limited to constant power only

• Very good harmonic

Current Injection Techniques Current Controllers Merits Demerits

the power quality as normal operation. Many current controllers are

You might also like