energies

Article
Adaptability Analysis of Fault Component Distance
Protection on Transmission Lines Connected to
Photovoltaic Power Stations
Yingyu Liang *, Guanjun Xu, Wenting Zha and Cong Wang
School of Mechanical Electronic and Information Engineering, China University of Mining and
Technology (Beijing), Beijing 100083, China; [email protected] (G.X.);
[email protected] (W.Z.); [email protected] (C.W.)
* Correspondence: [email protected]; Tel.: +86-010-6233-1370




Received: 22 March 2019; Accepted: 23 April 2019; Published: 25 April 2019


Abstract: Photovoltaic (PV) power stations tend to have a relatively weak infeed characteristic, unlike
conventional synchronous generators. The limited overcurrent capability of power electronic devices
and the controllability of grid-connected inverters mean that PV power stations will cause changes in
the characteristics of faults on transmission lines. To analyze the adaptability of fault component
distance protection on transmission lines connected to PV power stations, a unified phasor expression
for the fault current of a PV power station side under various control strategies was deduced in this
paper. This expression is then used to derive the equivalent impedance on the PV power station side
and the additional impedance. The equivalent impedance and additional impedance are affected
greatly by the active and reactive power commands, control targets, and fault conditions. These
aspects of a PV power station may cause malfunctions, which can thereby reduce the reliability
of fault component distance protection on transmission lines connected to PV power stations. A
simulation model of a PV power station was established in PSCAD/EMTDC and the correctness of
theoretical analysis was verified by the simulation results.

Keywords: fault component distance protection; equivalent impedance; adaptability analysis; PV

power station

1. Introduction
With the rapid development of photovoltaic (PV) power-generation technology and the increasing
installed capacity of PV power stations, PV power generation has achieved a pivotal position in the
global energy system [1–3]. Unlike traditional power supply based on synchronous generators, PV
power stations have a high degree of controllability. The fault-current characteristics on the PV power
station side are affected by the adopted control strategy, control parameters, and fault conditions,
among other factors. Additionally, due to the weak infeed characteristic of the PV power station,
the fault current in a transmission line will have a lower magnitude than a fault with a conventional
power supply [4–6]. Therefore, the fault-current characteristics of a PV power station are significantly
different from that of a synchronous source. Therefore, the fault analysis and protection schemes used
in synchronous source are not suitable for the transmission line connected to a PV power station [7–9].
Fault component distance protection offers the advantages of clear directionality, high-speed operation,
and is not affected by the pre-fault load current, among others [8]. The fault component distance
protection is widely used in high-voltage transmission lines. Therefore, studying the influence of PV
power station on fault component distance protection has a great significance, and corresponding
solutions should be proposed to ensure the safe and stable operation of the power systems.

Energies 2019, 12, 1578; doi:10.3390/en12081578 www.mdpi.com/journal/energies

Energies 2019, 12, 1578 2 of 19

Several researches have addressed control strategies for photovoltaic systems but relatively few
papers investigated the effects of PV power stations on fault component distance protection. Based on
the fault characteristics of traditional power supply, a fault component distance protection is introduced
in [10]. Using the current characteristics of a thyristor-controlled series compensator (TCSC), the
impact of TCSC on fault component distance protection was analyzed, and it was shown that the fault
component distance protection may perform an incorrect operation [11]. A novel weak fault component
distance protection was proposed previously, which integrates the performance of a fault component
reactance relay and a fault component impedance relay by adjusting the compensated voltage [12].
Simulations showed that this protection scheme offers an improved sensitivity and selectivity while
responding to unbalanced faults. Fault component distance protection based on voltage amplitude
comparison was introduced and its merit in covering larger fault resistances was analyzed for unified
power flow controller (UPFC) [13]. References [10–13] analyzed fault component distance protection
but PV power stations differ from other power systems from the perspective of operating principles and
control strategies. Since PV power stations will be interconnected with the existing power networks,
the adaptability of the currently protection methods need to be further investigated.
Currently, distance protection is widely utilized in high-voltage transmission networks and
several studies have been conducted from the viewpoint of the impact of renewable energy power
stations on distance protection. The adaptability problem of typical relay protection schemes to
transmission lines emanating from PV power stations was briefly studied in [14]. The fault-current
characteristics within a large PV power station were analyzed in [15], and a distance protection scheme
was proposed to effectively isolate PV power station from faults in overhead transmission lines. In
particular, PV power stations have some impact on traditional current protection and may lead to
protection maloperation [16]. The effect of voltage source converter based high voltage direct current
(VSC-HVDC) interconnectors on distance protection of transmission lines has also been analyzed
in [17,18]. Additionally, a novel pilot protection principle was proposed in [17]. Furthermore, the
changes of modular multilevel converter based HVDC (MMC-HVDC) operation mode cause the
scope of higher line distance relay Zone 2 to be narrowed, considerably increasing the probability
of incorrect operation of distance protection [19]. The influence of full-scale converter-interfaced
renewable energy power plants (CIREPPs) on distance relays installed on transmission lines connected
to CIREPPs has been investigated in [20], and another novel protection scheme was proposed in [21].
The impact of the several control strategies and the combined impact of a voltage source converter
(VSC) with a synchronous condenser on distance protection has been evaluated in [22]. Results showed
that the control strategies should avoid the use of constant reactive power to enhance the reliability
of distance protection. In addition, wind farms based on doubly fed induction generators (DFIGs)
are widely employed in existing power networks, and numerous studies have been performed on
distance protection performance in wind farms [23–26]. Considering the specific fault characteristics
and complex control strategies of DFIGs in wind farms, a new adaptive distance protection scheme
was proposed in [23] and a distance relay in the time domain based on the resistance- inductance (R-L)
differential-equation algorithm was proposed in [26]. However, none of these studies discussed fault
component distance protection or considered specific PV power station characteristics. In particular,
unlike conventional distance protection, fault component distance protection is based on a voltage and
current fault component, which significantly differs in terms of operating principles and advantages.
Furthermore, PV power stations differ from DFIG-based wind power integration systems from the
perspective of operation mode and fault-current characteristics. Therefore, it is highly necessary to
investigate the adaptability problem of the fault component distance protection on the transmission
lines connected to PV power stations.
The rest of this paper is organized as follows. Section 2 formulates a phasor expression for the
fault current on the PV power station side and discusses the fault-current characteristics. Section 3
describes the general principles for fault component distance protection and uses the phasor expression
of fault current to derive the equivalent impedance of the PV power station side and the additional
 Energies 2019, 12, 1578 3 of 18
Energies 2019, 12, 1578 3 of 19

expression of fault current to derive the equivalent impedance of the PV power station side and the
additionalFurthermore,
impedance. impedance. Furthermore, this section
this section analyzes theanalyzes
influencethe
of influence
equivalent ofimpedance
equivalent impedance of
of the PV power
the PV power station side and the additional impedance on fault component distance protection.
station side and the additional impedance on fault component distance protection. The simulation The
simulation
results results
presented presented
in Section in Section
4 verify the 4theoretical
verify the analysis
theoretical
of analysis of the adaptability
the adaptability of fault
of fault component
component distance protection. Discussion and conclusions are drawn in Sections 5 and 6,
distance protection. Discussion and conclusions are drawn in Sections 5 and 6, respectively.
respectively.
2. Fault-Current Characteristics at a PV Power Station
2. Fault-Current Characteristics at a PV Power Station
A practical 150-MW PV power station connected to a power grid is used as the model for analysis,
A practical 150-MW PV power station connected to a power grid is used as the model for
as shown in Figure 1. The M side is the PV power station side and the P side is the system side. The PV
analysis, as shown in Figure 1. The M side is the PV power station side and the P side is the system
power station is connected to a grid with a two-stage boosting structure. First, the direct current (DC)
side. The PV power station is connected to a grid with a two-stage boosting structure. First, the direct
output of the PV power generation unit is inverted into a power frequency alternating current (AC)
current (DC) output of the PV power generation unit is inverted into a power frequency alternating
through
currentthe grid-connected
(AC) inverter, and is
through the grid-connected then integrated
inverter, and is theninintegrated
a 35 kV switch
in a 35cabinet.
kV switchFinally, after
cabinet.
35 kV busbar
Finally, afterconvergence,
35 kV busbar the power output
convergence, is boosted
the power outputand sent toand
is boosted thesent
local
to power
the localgrid through
power grid a
220through
kV double-winding transformer.
a 220 kV double-winding transformer.

PV power station
Inverter Main
transformer
N M Transmission P Ep(220kV)
Line
PV
35kV 220kV
f

Figure1.1.Model
Figure Modelof
ofaagrid-connected
grid-connected photovoltaic
photovoltaic(PV)
(PV)power
powerstation.
station.

When
When asymmetric
asymmetric faults
faults occurininthe
occur theAC
ACtransmission
transmissionline,
line, MP,
MP,the
the three-phase
three-phase voltage
voltage phasor
phasor of
of the PV power station (M side) can be expressed
the PV power station (M side) can be expressed as: as:

. U Mη. + ( ) (
= U η+ .+−U η− +.U0 0 = U + δ+ + θη +U − δ−− θη + k0U 0δ0 ,
UMη = Uη + Uη + U = U+ ] δ+ + θη + U− ] δ− − θη + k0 U0 ]δ0 ,
) (1)
(1)
where U is the voltage magnitude and δ is the initial phase angle of the voltage. The superscripts “+,”
“−,”U
where andis the
“0” voltage magnitude
denote positive, and δ is
negative, andthezero-sequence
initial phase components,angle of the voltage. The ηsuperscripts
respectively. = a, b, and c,“+,”
“−,”θaand
= 0°,“0”
θb =denote positive,
−120°, and negative,
θc = 120°; and k0 and
is thezero-sequence
grounding coefficient. components, respectively.
If the fault is a groundη =fault,
a, b,kand
0 is c,

θa = 0◦ , θtob 1;
equal = otherwise,
−120◦ , andk0θis c= 120◦to
equal 0. k0 is the grounding coefficient. If the fault is a ground fault, k0
; and
is equalTheto 1;fault-current
otherwise, kcharacteristics
0 is equal to 0.on the M side depend on the control system of a PV grid-
connected
The fault-current characteristicsunbalanced
inverter. Considering on the M side grid dependvoltage, on the a current controller
control system of aunder a double
PV grid-connected
synchronous rotating frame is widely used in PV grid-connected
inverter. Considering unbalanced grid voltage, a current controller under a double synchronous inverters [27,28]. Therefore, various
control
rotating targets
frame are set in
is widely PV grid-connected
used in PV grid-connected inverterinverters
control systems, [27,28]. and a unified
Therefore, expression
various controlfortargets
the
reference current under various control targets can be expressed:
are set in PV grid-connected inverter control systems, and a unified expression for the reference current
under various control targets can be expressed:  u+ P u+ Q 
 α ref 2 + β ref2 
1 − k+χ kρ 1 +uk+χQkref ρ 

 +uα Pref2 + β 2  
 +i +  Prefχ kρ uα+1Q
 uβ1−k +kχ kρ  
 iαref αref
 uβ+ P2ref− uα+ Qref
 + i + 
ref
 
 i  βref 2  1 − k χ kρ 2 − 1 + kχ kρ22  
2

 βref  =  1−kχ kρ 1+kχ kρ , 
 − −  =   ,

(2) (2)
 αref 
 i− i − 
( )
 i iαref 33(UU++2)2kχ ukχα−uP−αrefPref kχ ukχβ−uQ−β ref
 1−kχ2kρ2− − 1+kχ2kρ2 
Qref 
 

 βref 
βref 1 −kχkuχ−βkPρref 1 +k kuχ−kQρ  
χ α ref 
k u− P 2 + − 2
χ kρ k χ u1α+ Qkref

 χ 1−k β ref
+
χ kρ
 1 − k k2 1 + k k2 
where k is a real number that defines the control  targets χ ρ
during the
χ ρ 
fault-ride-through period. The
χ
subscripts
where kχ“ref” denotes
is a real thethat
number reference
defines value. When
the control kχ = during
targets −1, 0, and 1, the control targets
the fault-ride-through of the
period. The PV
grid-connected inverter
subscripts “ref” denotesare
theto eliminate
reference reactive
value. Whenpower
kχ = −1,oscillation,
0, and 1, theto eliminate
control negative-sequence
targets of the PV grid-
current, and to eliminate active power oscillation, respectively. Pref and Qref are the reference values of
Energies 2019, 12, 1578 4 of 19

active and reactive power commands, respectively, and kρ is the grid voltage unbalance coefficient
(kρ = U− /U+ ).
When asymmetrical faults occur in the transmission line, since the step-up transformer
(high-voltage side) is connected in star, the fault current in the transmission line comprises positive-,
negative-, and zero-sequence components. After the reference currents from the control strategy in (2)
have been applied, the phasor expression for steady-state fault current on the M side is as follows:
.    
IMη = Ivm ] δ+ + ϕ + θη − kχ kρ Ivm ] δ− + ϕ − θη + k0 I0 ]θ0 , (3)

where s !2 !2
2 Pref Qref
Ivm = 3U+
+
1−kχ kρ2 1+kχ kρ2
, (4)
−Qref /(1+kχ kρ2 )
ϕ= tan−1
Pref /(1−kχ kρ2 )

where Ivm is the magnitude of the positive-sequence current component. I0 and θ0 represent the
magnitude and initial phase angle of the zero-sequence current, respectively. The magnitude and initial
phase angle are determined by the zero-sequence voltage and the impedance of the zero-sequence
network, respectively.
For synchronous sources, the fault current of traditional synchronous sources has been well
studied, and its expression can be expressed as [29,30]:

√ x0 d − x00 d T00−t xd − x0 d T−t0

!
 
iη−SG = 2Eq|0| e d + e d cos ωt + ϕη
x0 d x00 d xd x0 d
√ √ (5)
2Eq|0|   2Eq|0| cos ϕη −t
+ cos ωt + ϕη − e Ta , t ≥ t0
xd x00 d

where η is phase A, B, or C, and SG denotes the synchronous generator. Eq|0| and xd are the inner
potential and reactance of the generator, respectively. x0 d and x00 d are the transient reactance and
sub-transient reactance, respectively. T0 d and T00 d are the transient time constant and sub-transient time
constant, respectively. ω is the power frequency electrical angular velocity, ϕη is the fault inception
angle, Ta is the time constant of the stator windings, and t0 is the fault inception time.
As seen in Equation (5), the fault current comprises an exponentially decaying cosine wave, a
steady cosine wave, and an exponentially decaying DC component. The fault-current characteristics of
a synchronous source are primarily determined by the synchronous source’s characteristics and fault
conditions. Generally, the fault-current magnitude is much larger than the rated value.
According to Equations (3)–(5) and the vulnerability of power electronic devices, the observed
characteristics are summarized as follows:

• The fault-current characteristics on the M side differ significantly from those of a synchronous
source and are affected by a wide range of factors, such as the control targets, active and
reactive power commands, and the voltage unbalance on the grid. The positive-sequence and
negative-sequence currents of M side are completely controllable, whereas fault currents of the
synchronous source are uncontrollable, which is depended on synchronous source characteristics
and fault conditions.
• Because of the overcurrent limits of power electronic devices and the controllability of grid-facing
inverters, the maximum magnitude of positive and negative-sequence currents in a fault current
generally do not exceed 1.2 times the rated value. However, the magnitude of the fault current of
the synchronous source is much larger than the rated value of the normal current.
• The main transformers usually adopt YNd11 wiring and PV power stations tend to have a weak
infeed for the positive- and negative-sequence components. When a ground fault occurs in the
transmission line, the fault current consists of positive, negative, and zero-sequence components,
Energies 2019, 12, 1578 5 of 19

and the zero-sequence current dominates. When an ungrounded fault occurs, the fault current
does not include a zero-sequence component.

3. Adaptability Analysis of Fault Component Distance Protection

Based on the analysis of fault current of transmission lines in Section 2, the adaptability of typical
fault component distance protection on the M side are analyzed in detail below.

3.1. Principle of Fault Component Distance Protection

The usual operation criteria for fault component distance protection is
 .   . .   . 
∆Uop  = ∆UM − ∆IM Zset  ≥ U[0]  internal fault
 f 
 .   . . . (6)
∆Uop  = ∆UM − ∆IM Zset  < U[0] 
  
 f  external fault

. . [0]
where ∆Uop is the fault component of a compensated voltage and U f is the pre-fault voltage at fault
. .
point. ∆UM and ∆IM are the fault components of the voltage and current at relay point M, respectively.
Zone 1 is taken as an example to study the effect of PV power station on fault component distance
protection in transmission lines. The analysis method for Zone 1 and the conclusion obtained from
Zone 1 are applicable to Zone 2. Zset is the setting impedance of Zone 1 (i.e., 80% of the line impedance).
The operating principle for traditional fault component distance protection is shown in Figure 2.
Since the impedance angles of various elements in fault-additional network for traditional power
. . . [0]
gridEnergies 1578 the phases of ∆U , ∆U , and U
2019, 12,equal,
are nearly 6 of 18
op M f are almost the same. In normal conditions,
.
∆Uop = 0; therefore, the fault component distance protection will  not engage. During an external
. . [0] ΔU Mη  U .  
M  . [0] 
metallic fault, ∆Uop and U f have the = − phase, and
Zs same , Zmas=∆Uop  < ,U f , the fault component distance
η
(9)
ΔIMη + kl I0 IMη + kl I0  .   . [0] 
protection will still not operate reliably. For the internal metallic fault, ∆Uop  ≥ U f ; hence, the fault
 
where Zs is the equivalent impedance on the M side and Zm is the measured impedance. η = a, b, and
component distance protection can operate successfully. For a fault at the end of the protection zone,
c, φφ = ab, bc, and ca. kl is the zero-sequence compensation coefficient, kl = (z0−z1)/(z1), and z1 and z0
theare
fault
thecomponent distance
positive sequence andprotection is on impedance
zero-sequence the edge ofofoperation.
unit length, respectively.

jX
~P ~P
f2 U [0]
ΔU op
f
Zset
ΔU op
Zm f1 U [0]
f

Zm
Zset Zm Zset
R
Zs ΔIM ΔU M ΔIM ΔU M

Zs Zs
~M ~M
(a) (b) (c)
Figure
Figure 2. Operating
2. Operating principleofoffault
principle faultcomponent
component distance
distance protection:
protection:(a)(a)Operating
Operatingperformance
performanceof of
fault component distance protection; (b) internal metallic faults; and (c) external metallic
fault component distance protection; (b) internal metallic faults; and (c) external metallic faults. faults.

Generally, the setting impedance is not affected by PV power station connected to the power
grid, and the value of Zs is basically unchanged in traditional power grid. However, according to the
phasor expression in (3) and (4), the fault-current characteristics on the M side differ significantly
from those of conventional power grid. Therefore, Zs and Zm are affected greatly by the
interconnection of PV power station, which lead Zs and Zm to differ significantly from their values in
traditional power grid in terms of magnitude and phase angle, and may further lead fault component
distance protection to operate incorrectly.

3.2. Characteristics of Equivalent Impedance on the PV Power Station Side

Energies 2019, 12, 1578 6 of 19

According to (6) and Figure 2, when a fault occurs in the transmission line, the impedance phase
angles of Zs , Zm , and Zset are equal, which is the ideal condition for correct operation of fault component
distance protection. Therefore, the operation criterion of (6) can be transformed into impedance form
as follows:
|Zs + Zset | ≥ |Zs + Zm | internal fault
, (7)
|Zs + Zset | < |Zs + Zm | external fault
In phase-to-phase faults, Zs and Zm are obtained as:
. .
∆UMϕϕ UMϕϕ
Zs = − . , Zm = . , (8)
∆IMϕϕ IMϕϕ

In a single-phase-to-ground fault, Zs and Zm are obtained as

. .
∆UMη UMη
Zs = − . . , Zm = . . , (9)
∆IMη + kl I0 IMη + kl I0

where Zs is the equivalent impedance on the M side and Zm is the measured impedance. η = a, b, and
c, ϕϕ = ab, bc, and ca. kl is the zero-sequence compensation coefficient, kl = (z0 − z1 )/(z1 ), and z1 and
z0 are the positive sequence and zero-sequence impedance of unit length, respectively.
Generally, the setting impedance is not affected by PV power station connected to the power grid,
and the value of Zs is basically unchanged in traditional power grid. However, according to the phasor
expression in (3) and (4), the fault-current characteristics on the M side differ significantly from those
of conventional power grid. Therefore, Zs and Zm are affected greatly by the interconnection of PV
power station, which lead Zs and Zm to differ significantly from their values in traditional power grid
in terms of magnitude and phase angle, and may further lead fault component distance protection to
operate incorrectly.

3.2. Characteristics of Equivalent Impedance on the PV Power Station Side

According to previous discussion, if the measured impedance is equal to the actual fault impedance,
the operating performance of fault component distance protection is determined by the equivalent
impedance characteristics of M side. Therefore, this section considers an internal metallic fault as
an example to study the equivalent impedance characteristics of the M side, and then analyzes the
adaptability of fault component distance protection. The influence of the phase angle change of Zs on
the operating performance of fault component distance protection is shown in Figure 3. The dotted
and solid lines illustrate the cases in which the magnitude of Zs is small and the magnitude of Zs is
large, respectively.
Figure 3 shows that the phase angle change of Zs affects the operating performance of fault
component distance protection. The following preliminary conclusions can be drawn concerning the
operating performance of fault component distance protection:

• When 0◦ < arg(Zs ) < 180◦ , the impedance relationship still meets the operation criterion
of impedance form in Equation (7). The fault component distance protection will operate
correctly; however, the large magnitude of Zs decreases the sensitivity of the fault component
distance protection.
• When −90◦ < arg(Zs ) < 0◦ and the magnitude of Zs is small, the impedance relationship still meets
the operation criterion of the impedance form. When the magnitude of Zs is large, the impedance
relationship does not satisfy the operation criterion of impedance form. Due to the weak infeed
characteristic of the PV power station, the magnitude of Zs is much larger than that in traditional
power grid. Therefore, the fault component distance protection will operate incorrectly.
Energies 2019,
Energies 12,12,
2019, 1578
1578 7 of 718of 19

characteristic of the PV power station, the magnitude of Zs is much larger than that in traditional
grid.<Therefore,
arg(Zs ) < −90
• When −180 ◦ ◦ , the impedance relationship does not satisfy the operation criterion
power the fault component distance protection will operate incorrectly.
• of impedance
When −180°form, and
< arg(Z s) <the fault
−90°, thecomponent distance protection
impedance relationship does notwill be more
satisfy sensitivecriterion
the operation to changes
of Z
ofs .impedance
Therefore, form,
the fault
andcomponent distance protection
the fault component will operate
distance protection incorrectly.
will be more sensitive to
changes of Zs. Therefore, the fault component distance protection will operate incorrectly.

jX jX
Zs + Zm Zs + Zm

Zs + Zset Zs + Zset

Zset Zset
Zs
Zm Zm
Zs
R R

(a) (b)

jX jX

Zset Zset

Zm Zm

R R

Zs Zs + Zm Zs Zs + Zm
Zs + Zset Zs + Zset

(c) (d)
Figure
3. 3. Influenceofofphase
Influence phaseangle
angle change
change of
of ZZs on the operating performance of fault component
Figure s on the operating performance of fault component
distance protection: (a) 0°◦ < arg(Zs) < 90°; (b) 90° < arg(Z s) < 180°; (c) −90° < arg(Zs) < 0°; and (d) −180°
distance protection: (a) 0 < arg(Zs ) < 90◦ ; (b) 90◦ < arg(Zs ) < 180◦ ; (c) −90◦ < arg(Zs ) < 0◦ ; and
< arg(Z s ) < −90°.
(d) −180◦ < arg(Zs ) < −90◦ .

3.2.1.
3.2.1. CharacteristicAnalysis
Characteristic AnalysisofofZZss under
under Phase-to-Phase
Phase-to-Phase Fault
Fault
When
When anan internalfault
internal faultoccurs
occursat
at the
the transmission
transmission line
line MP
MP(f),
(f ),the
thephasor
phasorexpressions
expressionsforfor
thethe
three-phase voltage and three-phase current of M side before the fault can be expressed as
three-phase voltage and three-phase current of M side before the fault can be expressed as follows: follows:

  [0] U +  +

. U (
[0] = U+  δ − Δδ + φ  )
 Mη = kkλλ ] δ − ∆δ + ϕη


 U Mη + η

.[0][0]
( )

IMη== IINm
  +
 δδ+ − δ ++φϕ , (10) (10)
 

 IMη Nm ] −Δ∆δ η η,



2k P0
INm = 23Ukλ λ+P



I Nm = 0
 3U +

where ∆δ is the jump angle of the positive-sequence voltage, kλ is the coefficient of the positive-sequence
wheresag,
voltage ∆δ and
is the
P0jump
is theangle
powerofoutput
the positive-sequence
of the PV power voltage, λ is the normal
stationkduring coefficient of the positive-
operation.
sequence voltage sag, and P0 is the power output of the PV power station during normal operation.
 Energies 2019, 12, 1578 8 of 18
Energies 2019, 12, 1578 8 of 19

PV power station usually adopts the control target of eliminating a negative-sequence current,
so kχ = 0. Assuming that an internal phase-to-phase fault occurs at f, if (3), (4), and (10) are substituted
PV power station usually adopts the control target of eliminating a negative-sequence current, so
into (8), the magnitude and phase angle of Zs can be calculated as
kχ = 0. Assuming that an internal phase-to-phase fault occurs at f, if (3), (4), and (10) are substituted
into (8), the magnitude and phase ΔU angle
Mφφ
of Zs can be calculated ΔU Mφφ as
Zs = = ,
 . ΔIMφφ  2 .
 I   U 2 I vm

(11)
 ∆UMϕϕ  3 I Nm 1 +  vm ∆  −Mϕϕ  cos(φ + Δδ)
|Zs | =  . = r  I Nm  I , (11)
 ∆IMϕϕ  √  Nm
Ivm 2

2Ivm
3INm 1 + I − I cos(ϕ + ∆δ)
 ΔU Mφφ   C+ Zs +Nm + C− Zs − Nm 
arg ( Zs ) =arg  −  =arg  = →
 ΔIMφφ  . ! C+ + C− 
 ∆UMϕϕ C+ Zs+ +C− Zs−

arg(Zs ) = arg − . = arg =→
  ∆IMϕϕ I vm  C I+
+ C−

 kλ sin Δδ +
 sin φ  − Ivm sin ( φ + Δδ ) , ,
Ivm
vm (12) (12)
k sin ∆δ + sin ϕ − sin ( ϕ + ∆δ )
−1   
−1 λ I NmINm  IINm Nm

→ tan→ tan 
vm cos(ϕ+ ∆δ)−1  
   
 k cos ∆δ− Ivm cos ϕ +IIIvm

 k  cosλΔδ − I vmINm cos φ  + (
Nm cos φ + Δδ − 1 
) 
 λ I Nm
 . 
   I Nm 
where ∆UMϕϕ  is the magnitude of the difference between the two phases of the voltage when a
 
Where ΔU Mφφfault
phase-to-phase occurs
is the (refer to
magnitude ofAppendix
the difference A Equation
between the (A1)),twoand phasesC+ of and − are coefficients
theCvoltage when a
describing
phase-to-phase fault occurs (refer to Appendix (A1)), and C and C are coefficients describing station
the distribution of the positive- and negative-sequence + current.
− When the PV power the
adopts the control
distribution of the target of eliminating
positive- the negative-sequence
and negative-sequence current. When current,the PV then powerC− =station
0. adopts the
Due to
control the weak
target infeed characteristic
of eliminating the negative-sequenceof PV power current, station,
then Cthe − = 0.magnitude of Zs is much larger
than that in a traditional power grid. According to
Due to the weak infeed characteristic of PV power station, the magnitude (4) and (12), arg(Z s ) is relatedof toZthe
s is variables
much larger of ∆δ,
P*,than
Q*, andthat kinλ .aThe difficulty
traditional power of theoretical
grid. According analysis to (4)increases
and (12),significantly
arg(Zs) is related owing to to
thethe presence
variables of of
several
∆δ, Pvariable
*, Q*, andfactors.
kλ. The When
difficulty theofthree-phase
theoretical fault analysis 0 Ω fault
withincreases resistance occurs,
significantly owing to = 0presence
kλthe (minimum
of several
value). variableasfactors.
kλ increases the fault When the three-phase
resistance increases.fault When withthe 0 value
Ω faultofresistance
fault resistance = 0
occurs,iskλinfinite,
kλ = 1 (maximum value). However, it is impossible that the value of fault resistance is infinite
(minimum value). k λ increases as the fault resistance increases. When the value of fault resistance is
infinite, kλ = 1 (maximum value). However, it is impossible
when a fault occurs. Three-phase metallic fault rarely occurs. According to simulation data, the that the value of fault resistance is infinite
when
range of a∆δ fault
is occurs.
generally Three-phase
between metallic
−25◦ and fault5◦ .rarely
In view occurs. According
of most possible to simulation
conditions, data, thefollowing
the range
of ∆δ is generally between −25° and 5°. In view of most possible
assumptions are applied: (1) the coefficient of the positive-sequence voltage sag: 0.3 ≤ kλ ≤ 0.8 and conditions, the following assumptions
are applied: (1) the coefficient of the positive-sequence voltage sag: 0.3 ≤ kλ ≤ 0.8 and (2) the positive-
(2) the positive-sequence voltage jump angle: −π/8 < ∆δ < 0.
sequence voltage jump angle: −π/8 < ∆δ < 0.
The arg(Zs ) with different kλ and ∆δ under the control target of eliminating negative-sequence
The arg(Zs) with different kλ and ∆δ under the control target of eliminating negative-sequence
current is shown in Figure 4, drawn using MATLAB/2016B. Figure 4 shows that the arg(Zs ) lies between
current is shown in Figure 4, drawn using MATLAB/2016B. Figure 4 shows that the arg(Zs) lies
−80◦ and 110◦ under these conditions. Furthermore, as derived in Figure 3, when −180◦ < arg(Zs ) < 0◦ ,
between −80° and 110° under these conditions. Furthermore, as derived in Figure 3, when −180° <
thearg(Z
faults)component
< 0°, the fault distance
component protection
distance will fail to operate.
protection will fail to operate.

150
100

100
50
50
arg(Zs)/(°)
0 0

-50
-50
-100
0
-π/32 0.8
-π/16 0.7
0.6
Δδ -3π/32 0.5 k
-π/8 0.4 λ
0.3
Figure
Figure 4.4.The
Thecharacteristics
characteristics of
of arg(Z
arg(Zss))with
withdifferent
differentkλkλand ∆δ.∆δ.
and
Energies 2019, 12, 1578 9 of 19

In summary, when an internal phase-to-phase metallic fault occurs at f, the magnitude of Zs will
be large, and the phase angle is related to the active and reactive power commands, control targets,
and fault conditions. Therefore, when −180◦ < arg(Zs ) < 0◦ , the fault component distance protection
will fail to operate.

3.2.2. Characteristics of Zs under a Single-Phase-to-Ground Fault

When an internal phase-A-to-ground metallic fault occurs at f, by substituting (3), (4), and (10)
into (9), the magnitude and phase angle of Zs is calculated as
 . 
∆UMa 
|Zs | = s , (13)
(Ivm sin(δ+ + ϕ) − INm sin(δ+ − ∆δ) + (kl + 1)I0 sin θ0 )2 +
(Ivm cos(δ+ + ϕ) − INm cos(δ+ − ∆δ) + (kl + 1)I0 cos θ0 )2

(kl +1)I0
 
Ivm

INm sin(δ+ + ϕ) − sin(δ+ − ∆δ) + INm sin θ0 
arg(Zs ) = ∆θua − tan−1 
 , (14)
Ivm (kl +1)I0 

INm cos(δ+ + ϕ) − cos(δ+ − ∆δ) + INm cos θ0 
 . 
where ∆UMa  and ∆θua are the magnitude and phase angle of the A-phase voltage fault component
during the phase-A-to-ground fault, respectively (refer to Appendix A (Equations (A2) and (A3)).
As ∆θua is unaffected by the PV power station, and Ivm /INm ≈ 1.2 and I0 /INm > 3, the fault current
is dominated by the zero-sequence current. Therefore, ∆θia ≈ θ0 and 0◦ < arg(Zs ) < 180◦ . Generally,
owing to the weak infeed characteristic of PV power station, the magnitude of Zs in the PV power
station side is larger than that in a traditional power grid. As derived in Figure 3, when an internal
phase-A-to-ground metallic fault occurs at f, the fault component distance protection operates reliably.

3.3. Characteristics of Measured Impedance

Above, the equivalent impedance on the M side was analyzed theoretically, and the influence
of Zs on the fault component distance protection was considered under the assumption that Zm is
equal to the actual fault impedance (Zk ). The PV power station’s effect on Zm is considered below and
further analysis of the influence of Zm on fault component distance protection is given.
When a fault with fault resistance Rf occurs at f, the measured impedance (Zm ) will be:
. .
∆I M + ∆I P
Zm = Zk + ∆Z = Zk + . Rf , (15)
∆I M
. .
where ∆IM and ∆IP are the fault-current components on M and P side, respectively, Zk is the actual fault
impedance (positive-sequence line impedance between the protection location and the fault location f ),
and ∆Z is the additional impedance caused by the fault resistance.
For the internal metallic fault, the measured impedance will be equal to the actual fault impedance.
If the fault resistance Rf is non-zero, ∆Z will appear in the measured impedance, which means that the
distance relay will unable to calculate the actual fault impedance accurately. The characteristic of ∆Z
will differ for different faults; the following classification discusses the characteristics of ∆Z.
Energies 2019, 12, 1578 10 of 19

3.3.1. Characteristic Analysis of ∆Z under a Phase-to-Phase Fault

This section takes an internal phase B-to-phase C (BC) fault as an example to study the
characteristics of ∆Z and then analyzes the influence of PV power station on the measured impedance.
In this case, ∆Z can be expressed as
 . 
∆IPBC 
∆Z= R f + Z f = R f + R f  .  ](∆θPBC − ∆θMBC ), (16)
∆IMBC 

 .   . 
where ∆IPBC  and ∆IMBC  are the magnitudes of the difference between the BC-phase fault-current
components on the P side and M side, respectively. And ∆θMBC are the phase angles of the difference
between the BC-phase fault-current components on the P and M side, respectively.
. .
Generally, ∆IPBC is determined by the characteristics of traditional power supply, and ∆IMBC is
.
affected by the control parameters and fault conditions. From (3), (4), and (10), the value of ∆IMBC can
be calculated as
v
u
u  Ivm 2
 INm cos(δ + ϕ + θBC ) − cos(δ − ∆δ + θBC ) 
+ +
u
u
u
u
u
u
u
− IIvm kχ kρ cos(δ− + ϕ − θBC )
u
√
 .  u
u  
u
∆IMBC  = 3INm u u
u
t  Nm
2 , (17)
 IIvm sin(δ+ + ϕ + θBC ) − sin(δ+ − ∆δ + θBC ) 
+ Nm 
− IIvm kχ kρ sin(δ− + ϕ − θBC ) 
Nm

 Ivm sin(δ+ + ϕ + θBC ) − INm sin(δ+ − ∆δ + θBC ) 

 

−Ivm kχ kρ sin(δ− + ϕ − θBC )

 
 
−1 
∆θMBC = tan  , (18)
 
 Ivm cos(δ+ + ϕ + θ ) − I cos ( δ + − ∆δ + θ )  
 BC Nm BC  
−Ivm kχ kρ cos(δ− + ϕ − θBC )

 .   . 
where θBC = 3π/2·∆IPBC /∆IMBC  is large because of the weak infeed characteristic of PV power station.
In addition, the characteristics of ∆Z are mainly determined by ∆θMBC . According to Equations (4)
and (18), ∆θMBC values are affected by several factors, such as the control targets, active and reactive
power commands, and voltage unbalance on the grid. Consider a phase-to-phase fault as an example.
When the value of fault resistance Rf equals 0, kρ = 1 (maximum value). When the value of fault
resistance Rf is infinite, kρ = 0 (minimum value). However, the value of fault resistance is not zero,
and it cannot be infinite. Therefore, the range of kρ is between 0 and 1. Generally, the Pref value
during the fault is less than 1.0 pu and is related to the positive-sequence voltage sag. Therefore, to
simplify the analysis, the following assumptions are applied: (1) the active power reference values:
0.2 p.u. ≤ Pref ≤ 0.8 p.u.; and (2) the grid voltage unbalance coefficient: 0.2 p.u. < kρ < 0.8 p.u. The Pref
and kρ are within a reasonable range, and the range of Pref and kρ can cover most cases.
The ∆θBCM with different Pref and kρ under the control target of eliminating active power
oscillations is shown in Figure 5, which was drawn using MATLAB/2016B. Figure 5 shows that
the range of ∆θBCM is between −160◦ and −60◦ . The difference in phase angle between both sides
may express ∆Z as a capacitive or inductive impedance, which significantly differs from those of
a conventional power grid. Therefore, the controllability of ∆θBCM may causes the distance relay
to be unable to accurately calculate the actual fault impedance. The impedance plane of measured
impedance for an internal BC fault is shown in Figure 6a. The characteristics of Zs with different
influence factors.
Energies 2019, 12, 1578 11 of 19
Energies 2019,
Energies 2019, 12,
12, 1578
1578 11 of
11 of 18
18

-40
-40
-60
-60 -60
-60
-80
-80
-80
-80
-100
-100
ΔθMMBC
BC (°)
-100
-100
-120
-120
-120
-120
-140
-140
-140
-140
-160
-160
-160
-160
-180
-180
0.2
0.2 0.8
0.8
0.4
0.4 0.6
0.6
0.6 0.4 k ρ 0.6
0.4
Pref
ref 0.8
0.8 0.2
0.2 ρ

Figure
Figure
Figure 5.The
5. 5. Thecharacteristics
The characteristicsof
characteristics of ∆θ
of ∆∆θθBCM with
BCM with
BCM different
withdifferent active
differentactive power
activepower commands
power (P ref
commands
commands (P ref))ref
(P and
and thethe
) and
the grid
grid
grid
voltage
voltage
voltage unbalance
unbalance
unbalance coefficient(k(k
coefficient
coefficient (k
ρ ρ
ρ )
)
). ..

jX
jX jX
jX
Zff 22

ZmBC2
Z ΔZ Zset
mBC2ΔZ22 Z Zset
Z set
set
Rff
R ΔZ
ΔZ
Zkk
Z Zkk
ΔZ11 Z ZmAG
ΔZ Z ff11 Z mAG

R
R R
R
ZmBC1
Z mBC1

((aa)) ((bb))
Figure
Figure
Figure 6. 6.6. Impedanceplane
Impedance
Impedance planeof
plane ofmeasured
of measured impedance:
measured impedance: (a)
impedance: (a) Impedance
(a)Impedance plane
Impedanceplane ofof
of
plane measured
measured
measuredimpedance
impedance
impedance
(Z
(Z mBC)) for
for an
an internal
internal phase
phase B-to-phase
B-to-phase C
C (BC)
(BC) fault;
fault; and
and (b)
(b) impedance
impedance plane
plane of
of measured
measured
(ZmBC ) for an internal phase B-to-phase C (BC) fault; and (b) impedance plane of measured impedance
mBC impedance
impedance
(ZmAG
(Z
(ZmAGmAG)) for
) forforananinternal
an internalphase-A-to-ground
internal phase-A-to-ground (AG)
phase-A-to-ground (AG) fault.
(AG)fault.
fault.

3.3.2.
3.3.2. CharacteristicAnalysis
Characteristic
3.3.2. Characteristic Analysis of∆Z
Analysisof
of ∆Z under
∆Z under aa Single-Phase-to-Ground
Single-Phase-to-Ground Fault
Single-Phase-to-GroundFault
Fault
When
When
When kχ kis is zero
kχχ zero
is zero and
andand with
withwith the weak
the
the weak weak infeed
infeed
infeed characteristic
characteristic
characteristic ofpower
of PVof PV power
PV power station,
station,
station, the positive-
the positive-
the positive-sequence
sequence
sequence
current current
current
is less is
than isthe less
less than the zero-sequence
than the zero-sequence
zero-sequence current and current
current and can be
andneglected.
can be neglected.
can be neglected. In this case,
In thisthe
In this case, case, the zero-
the zero-
zero-sequence
sequence compensation
sequence compensation coefficient
coefficient kcoefficient can be
kkll can be assumed
assumed toto be
be aa real
real number.
number. In
In this
this case,
case, ∆Z
∆Z can
can be
be
compensation l can be assumed to be a real number. In this case, ∆Z can be expressed as:
expressed as:
expressed as:
. . . . 
∆IΔ II +++∆
ΔMa ΔIIIPa 3R
3R
3R f   IM0 +
f  I I M0 ++ IIP0 IP0 
∆ZΔ
Δ=Z RR
Z == R
f Ma
Ma
.
Δ Pa
Pa ≈
≈ 1 +f k M0
≈ . P0  ,,  , (19) (19)
(19)
ff
∆Δ IIMa
IMa
Δ Ma 11++ kkll l IIM0
I 
M0M0 

. .
Where
Where
where IM0 IIand and
M0 and
M0 IIP0
IP0 are arezero-sequence
are
P0 the the zero-sequence
the zero-sequence currents
currents
currents measured
measured
measured at the
at
at the the
MM M and
and
and the
thethe PP side.
P side.side. Assumingthat
Assuming
Assuming
. .
thethat
zero-sequence
that the zero-sequence
the impedance
zero-sequence angles
impedance
impedance on the
angles
angles onM
on and
the
the MP
M side
and
and are equal,
PP side
side are equal,
are equal, IIM0
IM0 is in phase
M0 isis in withwith
in phase
phase with IIP0
IP0 , and P0 ,,
the
zero-sequence
and the
and current magnitude
the zero-sequence
zero-sequence on the on
current magnitude
current magnitude M side
on the Misside
the M smaller
side thanthan
is smaller
is smaller thatthat
than on the
that P side.
on the
on the side.Therefore,
PP side. Therefore,the
Therefore,
additional
the impedance
additional impedanceyields a large
yields aresistance
large value.
resistance The
value.impedance
The
the additional impedance yields a large resistance value. The impedance plane of measured plane
impedance of measured
plane of impedance
measured
forimpedance
an internal
impedance for AG
for anfault
an internal is shown
internal AG in
AG fault Figure
fault is
is shown 6b.
shown inin Figure
Figure 6b.
6b.

3.3.3. Influence
3.3.3.
3.3.3. ofof
Influence
Influence ofthe
theMeasured
the MeasuredImpedance
Measured on Fault
Impedance on
Impedance FaultComponent
Fault ComponentDistance
Component Distance
Distance Protection
Protection
Protection
According
According
According to to
the discussion
to the
the in the
discussion
discussion previous
in the
in the section,
previous
previous the the
section,
section, measured impedance
the measured
measured waswas
impedance
impedance affected
was greatly
affected
affected
greatly
bygreatly by
the fault the fault
resistance resistance
and and
fault fault
current, current,
which which
make make
the the distance
distance relay relay
unableunable
to to calculate
calculate the
by the fault resistance and fault current, which make the distance relay unable to calculate the the
actual
fault impedance accurately. To simplify the analysis, this section considers an internal BC fault as an
 actual fault impedance accurately. To simplify the analysis, this section considers an internal BC fault
as an example to study the influence of the measured impedance on the operating performance of
fault component distance protection. In this case, the influence of the measured impedance on the
operating performance of fault component distance protection is shown in Figure 7.
As derived in Figure 3, when an internal metallic fault occurs at f and the 0° < arg(Zs) < 90°, the
Energies 2019, 12, 1578 12 of 19
fault component distance protection will operate reliably. However, when an internal fault with fault
resistance occurs at f, it can be seen from Figure 7 that the additional impedance (∆Z) will make the
distance relay unable to calculate the actual fault impedance accurately, which may enhance the
example to study the influence of the measured impedance on the operating performance of fault
adverse influence of Zs on fault component distance protection in case that the PV power station is
component distance protection. In this case, the influence of the measured impedance on the operating
connected to a power grid; further, it will lead fault component distance protection to operate
performance of fault component distance protection is shown in Figure 7.
incorrectly.

jX Zs + Zm Zs + Zset jX

Z mBC

Zset ΔZ Z
set

Zk Zk

ΔZ Zs Zs

R R
Z mBC

(a) (b)

Figure
Figure Influence
7. 7. Influenceofofthe
themeasured
measured impedance on the
impedance on the operating
operatingperformance
performance of of fault
fault component
component
distance protection:
distance (a) Additional
protection: impedance
(a) Additional ∆Z is ∆Z
impedance resistive-capacitive; and (b)and
is resistive-capacitive; additional impedance
(b) additional
∆Zimpedance
is resistive-inductive.
∆Z is resistive-inductive.

4. As derived Verification
Simulation in Figure 3, when an internal metallic fault occurs at f and the 0◦ < arg(Zs) < 90◦ , the
and Analysis
fault component distance protection will operate reliably. However, when an internal fault with fault
To verify the correctness of the theoretical analysis discussed above, a 150-MW PV power
resistance occurs at f, it can be seen from Figure 7 that the additional impedance (∆Z) will make the
station, as shown in Figure 1, was modeled in PSCAD/EMTDC. The rated power of the PV station is
distance relay unable to calculate the actual fault impedance accurately, which may enhance the adverse
SPV = 150 MVA, rated capacity of the main transformer is ST = 200 MVA, the rated change ratio is kT =
influence of Zs on fault component distance protection in case that the PV power station is connected
230/37 kV, the connection type of main transformers is YNd11, and the short-circuit impedance is
to a16%.
power
Thegrid;
ratedfurther,
voltageitof
will
thelead fault component
transmission distance
line is 220 protection
kV, the to operate
line length incorrectly.
is lMP = 50 km, and the
positive-sequence and zero-sequence impedance parameters of the unit length of the line are
4. Simulation Verification and Analysis
0.107+j0.427 and 0.535+j1.153 Ω/km, respectively. The equivalent positive-sequence impedance and
To verify the
zero-sequence correctness
impedance ofpower
of the the theoretical
grid systemanalysis discussed
are 0.6+j6.972 above, aΩ,150-MW
and 0.8+j10.95 PV power
respectively.
To prevent overcurrent from damaging power electronic switching devices
station, as shown in Figure 1, was modeled in PSCAD/EMTDC. The rated power of the PV stationand improve the
is SPV = 150 MVA, rated capacity of the main transformer is ST = 200 MVA, the rated change ratio
safety of inverter operation, the actual PV power station usually adopts the control target of is
kT eliminating
= 230/37 kV,negative-sequence
the connection type current. Therefore,
of main the simulations
transformers is YNd11, discussed below focusimpedance
and the short-circuit on the
theoretical analysis of fault component distance protection based on the control target of
is 16%. The rated voltage of the transmission line is 220 kV, the line length is lMP = 50 km, and eliminating
thenegative-sequence
positive-sequence current.
and zero-sequence impedance parameters of the unit length of the line are
0.107 + j0.427 and 0.535 + j1.153 Ω/km, respectively. The equivalent positive-sequence impedance and
4.1. Simulation Verification of Fault-Current Characteristics on PV Power Station Side
zero-sequence impedance of the power grid system are 0.6 + j6.972 and 0.8 + j10.95 Ω, respectively.
ToFigure 8 shows
prevent the simulation
overcurrent results ofpower
from damaging variouselectronic
faults occurring at 40%
switching of theand
devices length of thethe
improve
transmission line from the relay located on the M side (f). The red, black, and
safety of inverter operation, the actual PV power station usually adopts the control targetgreen curves plot the of
A, B, and C phases, respectively. For fault-current waveforms of the AG fault in
eliminating negative-sequence current. Therefore, the simulations discussed below focus on theFigure 8a, the fault
current is dominated by the zero-sequence current, and the phase angles of the three-phase currents
theoretical analysis of fault component distance protection based on the control target of eliminating
are nearly equal. However, for the fault current waveforms of the BC fault in Figure 8b, the fault
negative-sequence current.
current is solely positive-sequence current, the three-phase current is symmetrical, and its magnitude
4.1. Simulation Verification of Fault-Current Characteristics on PV Power Station Side
Figure 8 shows the simulation results of various faults occurring at 40% of the length of the
transmission line from the relay located on the M side (f ). The red, black, and green curves plot the
A, B, and C phases, respectively. For fault-current waveforms of the AG fault in Figure 8a, the fault
current is dominated by the zero-sequence current, and the phase angles of the three-phase currents are
nearly equal. However, for the fault current waveforms of the BC fault in Figure 8b, the fault current is
solely positive-sequence current, the three-phase current is symmetrical, and its magnitude does not
Energies 2019, 12, 1578 13 of 19

Energies 2019, 12, 1578 13 of 18

exceed 1.2 times the rated value. These fault-current characteristics are consistent with the theoretical
does not exceed 1.2 times the rated value. These fault-current characteristics are consistent with the
analysis presented
theoretical in presented
analysis Section 2. in Section 2.

200

Voltage(kV) 0

-200

2
Current (kA)

-2

0.95 1 1.05 1.1

Time (s)
(a)

200
Voltage(kV)

-200

0.5
Current (kA)

-0.5

0.95 1 1.05 1.1

Time (s)
(b)
Figure
Figure 8. 8.
TheThe simulationresults
simulation resultsof
ofvarious
various faults
faults occurring
occurringatat40%
40%ofofthe
thelength
lengthof of
thethe
transmission
transmission
lineline from
from thethe relay
relay locatedon
located onthe
theM Mside:
side: (a)
(a) The
The waveforms
waveformsofoffault
faultcurrent
current and
andfault
faultvoltage under
voltage under
AG fault; and (b) the waveforms of fault current and fault voltage under
AG fault; and (b) the waveforms of fault current and fault voltage under BC fault. BC fault.

4.2.4.2. Simulation
Simulation VerificationofofEquivalent
Verification EquivalentImpedance
Impedance Characteristics
Characteristicson
onPV
PVPower
PowerStation Side
Station Side
Figure
Figure 9 shows
9 shows thecharacteristics
the characteristicsof of Z
Zss with
with different
differentinfluence
influencefactors
factorsunder
underBCBCfault at point
fault f. f.
at point
The characteristics of Z s with different fault resistances are shown in Figure 9a. As shown in
The characteristics of Zs with different fault resistances are shown in Figure 9a. As shown in
Figure 9a,9a,
Figure thethe magnitude
magnitude of Zofs is is large
Zslarge in cases,
in all all cases,
andand increases
increases as fault
as the the fault resistance
resistance increases.
increases. When
When Rf = 0, the phase angle of Z◦s is −76°. Upon comparing different values of Rf, one can see that
Rf = 0, the phase angle of Zs is −76 . Upon comparing different values of Rf , one can see that arg(Zs )
arg(Zs) also increases as the fault resistance increases. Meanwhile, arg(Zs) is converted from the range
also increases as the fault resistance increases. Meanwhile, arg(Zs ) is converted from the range of −90◦
of◦ −90°◦to 0° to 0° to 90°. This conversion means that the fault component distance protection is very
to 0 to 0 to 90◦ . This conversion means that the fault component distance protection is very likely
likely to malfunction.
to malfunction.
The characteristics of Zs with different reactive power commands (Qref) are shown in Figure 9b.
The characteristics
If Qref is of Zs with
small or negative, different reactive
the −180°<arg(Z power commands (Qref ) are shown
s)<0°, and then the fault component distanceinprotection
Figure 9b. If
Qrefwill
is small or incorrectly;
operate negative, the −180
if Q
◦ < arg(Zs ) <0°
ref is positive, the
◦
0 <, arg(Z
and then the fault component distance protection will
s) < 180°, and the distance-relay fault protection
operate incorrectly; if Q is positive, the 0 ◦ < arg(Z ) < 180◦ , and the distance-relay fault protection
will operate correctly. ref s
will operate correctly.
Energies 2019, 12, 1578 14 of 19
Energies
Energies 2019,
2019, 12,12, 1578
1578 1414ofof1818

4 4 100
100
RfRf=0Ω
=0Ω RRf=5Ω
f=5Ω R
Rff=0Ω
=0Ω RRf=5Ω
f=5Ω
R
Rff=1Ω RRf=10Ω
RfRf=1Ω
=1Ω RRf=10Ω
f=10Ω 50 =1Ω f=10Ω

s)/(°)
50

s)/(°)
Zs/(kΩ)
Zs/(kΩ)

arg(Z
arg(Z
2.5 0
2.5 0

-50
-50
1
1 1.1 -100
1.15 1.2 -1001.1 1.15 1.2
1.1 1.15 (s)
Time 1.2 1.1 1.15 (s)
Time 1.2
Time (s) Time (s)
(a)
(a)
2 120
2 Qref= −0.2pu Qref=0.0pu Qref=0.2pu 120 Qref= −0.2pu Qref=0.0pu Qref=0.2pu
Qref= −0.2pu Qref=0.0pu Qref=0.2pu Qref= −0.2pu Qref=0.0pu Qref=0.2pu
60

s)/(°)
60

s)/(°)
Zs/(kΩ)

arg(Z
Zs/(kΩ)

1 0

arg(Z
1 0
-60
-60
0
-120
0 1.1 1.15 1.2 1.1 1.15 1.2
Time -120 Time (s)
1.1 1.15 (s) 1.2 1.1 1.15 1.2
Time (s) (b) Time (s)

Figure 9. The characteristics

)
(binfluence
Figure 9. The characteristics ofofZsZwith
s withdifferent
different influence factors
factorsunder
underBCBCfault at point
fault f: (a)f :The
at point (a) The
characteristics
Figure 9. Theof
characteristics Zofs Zs with different
characteristics
with different fault
of Zs with
fault resistances;
different and
and(b)
influence
resistances; the
thecharacteristics
factors
(b) under BC fault
characteristics of Z
at
of with
s point different
f: (a)different
Zs with The
reactive power
characteristics of commands.
Zs with different fault resistances; and (b) the characteristics of Zs with different
reactive power commands.
reactive power commands.
4.3. Simulation
4.3. Simulation Verification
Verification of MeasuredImpedance
of Measured Impedance Characteristics
Characteristics
4.3. Simulation
Figure 10Verification
shows of Measuredresults
Impedance Characteristics
Figure 10 shows thethe simulation
simulation ofthe
results of the measured impedance
measured impedance (Z(Z
m) under unbalanced faults
m ) under unbalanced faults
occurring
Figure at f. Figure 10a shows the simulation results of Z m under different metallic faults, and
occurring at f.10Figure
shows the 10asimulation
shows the results of the measured
simulation results ofimpedance
Zm under(Zdifferent
m) under unbalanced faultsand
metallic faults,
Figure 10b shows the simulation results of Zm under unbalanced faults with 5-Ω fault resistance.
occurring at f. Figure 10a shows the simulation results of Z under
Figure 10b shows the simulation results of Zm under unbalanced faults with 5-Ω fault resistance.
m different metallic faults, and
As shown in Figure 10a,b, when a metallic fault occurs, the distance relay can calculate the actual
Figure 10b
As shown shows
in Figurethe simulation
10a,b, when results of Z m under unbalanced faults with 5-Ω fault resistance.
fault impedance accurately. When afaults
metallic
withfault
5-Ω occurs, the distance
fault resistance occur,relay can calculate
influenced by the the
faultactual
As shownaccurately.
fault current
impedance in Figure 10a,b,When when a metallic
faults with fault
5-Ω occurs,
fault the distance
resistance occur, relay can calculate
influenced by the the actual
fault current
of the M side, the measured impedance is much larger than the actual fault impedance, and
fault impedance accurately. When faults with 5-Ω fault resistance occur, influenced by the fault
of thetheMdistance
side, the relaymeasured impedance
will be unable is much
to calculate larger
actual than
fault the actual
impedance fault impedance,
accurately. Therefore, and
the the
current of the M side, the measured impedance is much larger than the actual fault impedance, and
distance relay will
impedance be unable
measured by theto calculate actual
distance relay on fault impedance line
the transmission accurately.
will fall Therefore,
outside the the impedance
protection
the distance relay will be unable to calculate actual fault impedance accurately. Therefore, the
zone, as
measured byshown
the distancein Figure 10c,d.
relay onWhen a phase-to-phase
the transmission linefault
willoccurs, Zm will
fall outside theinclude a largezone,
protection imaginary
as shown
impedance
component, measured
and the by the distance
measured relaywill
impedance on the transmission
differ greatly fromline
the will
actual fall outside
fault the protection
impedance.
in Figure 10c,d. When a phase-to-phase fault occurs, Zm will include a large imaginary component,
zone, as shown in Figure 10c,d. When a phase-to-phase fault occurs, Zm will include a large imaginary
and the measured
component, and30 impedance
the measured will differ greatly
impedance from30
will differ the actual
greatly fromfault impedance.
the actual fault impedance.
25 25
30 30 Impedance circle
20 20
25 25
Reactance(Ω)
Reactance(Ω)

15 15 Impedance circle
20 20
10 10
Reactance(Ω)
Reactance(Ω)

15 15
5 5
10 10
0 0
Impedance circle
5 -5 Zm 5 Zm
-5
AG AB
0 0 BG 00
Impedance circle BC
-5-15 Zm CG -5 ZmCA
AG
-10 0 10 20 -15
-10 AB 0 10 20
0 BG Resistance(Ω) 0 BC Resistance(Ω)
CG (a) CA
-15 -10 0 10 20 -15
-10 0 10 20
Resistance(Ω) Resistance(Ω)
(a)

Figure 10. Cont.

Energies 2019, 12, 1578 15 of 19
Energies 2019, 12, 1578 15 of 18

Energies 2019, 12, 1578

40 40 15 of 18

40
30 40
30
Impedance circle
30
20 30
20
Reactance(Ω) Impedance circle

Reactance(Ω)
Reactance(Ω) 20
10 20
10

Reactance(Ω)
10
0 10
0
-100 Impedance circle
-100
-10
-20 Zm Impedance circle
Zm -10
-20
AG AB
-20
-30 ZmBG -20
-30 ZmBC
AG
CG AB
CA
-30
-40 BG -30
-10 CG 0 10 20 30 -40 -10 BC 0 10 20 30
Resistance(Ω) CAResistance(Ω)
-40 -10 0 10 20 30) -40
(b -10 0 10 20 30
Resistance(Ω) Resistance(Ω)
Figure
Figure 10.10.The
Thesimulation
simulation results
results of
of the measured(impedance
the measured )
bimpedance (Z(Z
m) under unbalanced faults occurring
m ) under unbalanced faults occurring
at f:
at fFigure(a)
: (a) The The simulation results of Z m under different metallic faults; and (b) the simulation results of
10. simulation results
The simulation of Zofmthe
results under different
measured metallic(Zfaults;
impedance andunbalanced
m) under (b) the simulation results of
faults occurring
ZmZatmunder
underunbalanced
unbalanced faults
f: (a) The
faults with
with 5-Ω
simulation results of 5-Ω
fault
faultresistance.
Zm under resistance.
different metallic faults; and (b) the simulation results of
Zm under unbalanced faults with 5-Ω fault resistance.
4.4.
4.4. SimulationVerification
Simulation VerificationofofOperating
Operating Performance
PerformanceofofFault
FaultComponent
ComponentDistance
DistanceProtection
Protection
4.4. Simulation
Figure
Figure 1111 Verification
shows
shows thethe ofoperating
Operating
operating Performance
performance
performance of of ofFault
fault faultComponent
component
component Distance Protection
distance
distance protection
protection under under
various
various fault
fault conditions. conditions.
Figure 11 shows Figurethe Figure
11a,b 11a,b
presents
operating presents the
the operating
performance operating
of fault performance
performance
componentofdistance of fault
fault component component
distance
protection under
distance
various protection
protection an under
underconditions.
fault an
AG metallic AG fault
Figure metallic
11a,band fault
a BC
presents andmetallic
thea BC metallic fault, respectively.
fault, performance
operating respectively. ofFigureFigures
fault 11c,d11c,d
componentpresent
thepresent
distance
operatingthe operatingunder
protection
performance performance
anfault
of of fault component
AG metallic
component and a distance
faultdistance BC protection
metallic
protection fault,
under under an AGFigures
an AG faultfault
respectively. withwith Ω 5fault
511c,d
Ω fault resistance and a BC fault with 5 Ω fault resistance, respectively.
present and
resistance the operating with 5 Ω fault
a BC faultperformance of fault component
resistance, distance protection under an AG fault with 5
respectively.
Ω As As shown
faultshown
resistancein Figure
and a BC 11a, when
faultwhen an
with 5an AG
Ω AG metallic
fault metallic fault
resistance,fault occurs on line MP at location f, the 0° <
respectively.
in Figure 11a, occurs on line MP at location f, the
arg(Z s) < 90° and the fault component distance protection can operate successfully. Similarly, as
0 < arg(Zs ) < 90 and the fault component distance protection canonoperate
◦ As shown ◦ in Figure 11a, when an AG metallic fault occurs line MP at location f,Similarly,
successfully. the 0° < as
shown
arg(Z in Figure 11b, when a BC fault occurs on line MP at f, the −180° < arg(Z
s) < 90° and the fault component distance protection can operate ◦
s) < −90°, and the fault
successfully. Similarly, asthe
shown in Figure
component 11b, when
distance a BCwill
protection faultfail
occurs on lineinMP
to operate this f, theThese
atcase. < arg(Zs ) <of−90
−180 simulations ◦ and
the , fault
shown
fault in Figure
component distance 11b, when a BC
protection fault occurs on
will failwith line
to operate MP at f, the −180°
in this case. < arg(Z s) < −90°, and the fault
These simulations of the fault
component
componentdistance
distanceprotection
protection arewill
consistent
fail to operate the theoretical
in this case. analysis
Thesediscussed
simulations in Section
of the 3.fault
component distance
The results protection
shown in Figure are consistent
11c,d imply that with thethe theoretical
fault component analysis
distance discussed
protectionin will
Section
fail 3.
component distance protection are consistent with the theoretical analysis discussed in Section 3. to
The
operate results shown in Figure 11c,d imply that the fault component distance protection will fail
Theifresults
a faultshown
with fault resistance
in Figure occursthat
11c,d imply in the transmission
the fault component line. These protection
distance simulationswill of fail
fault
to
to component
operate if adistance
fault with fault
protection resistance occurs
are consistent within the transmission line. These simulations of fault
operate if a fault with fault resistance occurs in the
the theoretical
transmission results
line.discussed in Section of
These simulations 3. fault
component distance protection are consistent with the theoretical results discussed in Section 3.
component distance protection are consistent with the theoretical results discussed in Section 3.
335 220
Voltage(kV)

Voltage(kV)

220
335 ΔU op
U [0] ΔU op
Voltage(kV)

315 190
Voltage(kV)

U [0]
f
ΔU op
ΔU op
f
315 U [0] 190
U [0]
f
295 150 f
200 110
295 150
s)/(°)

s)/(°)

200
100 110
80
arg(Zarg(Z

arg(Zarg(Z
s)/(°)

s)/(°)

100
0 80
50

-1000 50
20
1 1.05 1.1 1.15 1.2 1 1.05 1.1 1.15 1.2
-100 Time (s) 20 Time (s)
1 1.05 1.1 1.15 1.2 1 1.05 1.1 1.15 1.2
(a)
Time (s)
(b)Time (s)
(a) (b)

Figure 11. Cont.

Energies 2019, 12, 1578 16 of 19
Energies 2019, 12, 1578 16 of 18

335 220
U [0]
f
U [0]
f

Voltage(kV)

Voltage(kV)
ΔU op
315 180 ΔU op

295 140
1 1.05 1.1 1.15 1.2 1 1.05 1.1 1.15 1.2
Time (s) Time (s)
(c) (d)
Figure
Figure 11.11.
TheThe operatingperformance
operating performance of
of fault
fault component
component distance
distanceprotection
protectionunder different
under fault
different fault
conditions:
conditions: (a)(a)
TheThe operatingperformance
operating performance of
of fault
fault component
componentdistance
distanceprotection
protectionunder
undera BC metallic
a BC metallic
fault;
fault; (b)(b)
thethe operatingperformance
operating performanceof
of fault
fault component
component distance
distanceprotection
protectionunder anan
under AGAG metallic
metallic
fault; (c) the operating performance of fault component distance protection under a BC fault with fault
fault; (c) the operating performance of fault component distance protection under a BC fault with fault
resistance; and (d) the operating performance of fault component distance protection under an AG
resistance; and (d) the operating performance of fault component distance protection under an AG
fault with fault resistance.
fault with fault resistance.
5. Discussion
5. Discussion
Currently,
Currently, somestudies
some studieshave
haveinvestigated
investigated the
the effects
effects of
of renewable
renewableenergy
energypower
power stations on on
stations
distance protection [14–26]. However, these studies have not discussed fault component distance
distance protection [14–26]. However, these studies have not discussed fault component distance
protection or considered the specific fault characteristics of PV power stations. Because of the
protection or considered the specific fault characteristics of PV power stations. Because of the increasing
increasing penetration of PV power generation, investigating the adaptability problem of fault
penetration of PV power generation, investigating the adaptability problem of fault component distance
component distance protection on transmission lines connected to PV power stations is crucial. The
protection on transmission lines connected to PV power stations is crucial. The main contributions of
main contributions of this paper are as follows：
this paper are as follows:
• The expression of equivalent impedance of a PV power station side is derived based on the
• Thephasor expression
expression of fault current.
of equivalent The equivalent
impedance impedance
of a PV power characteristic
station of the based
side is derived PV power
on the
station
phasor side is considerably
expression affectedThe
of fault current. by active and reactive
equivalent power
impedance commands, control
characteristic targets,
of the PV power
and fault conditions.
station side is considerably affected by active and reactive power commands, control targets, and
• fault
This paper studied the influence of phase angle change of Zs on fault component distance
conditions.
protection operating performance and found that additional impedance will enhance the
• This paper studied the influence of phase angle change of Zs on fault component distance
adverse influence of Zs on fault component distance protection in cases where a PV power
protection operating performance and found that additional impedance will enhance the adverse
station is connected to a power grid.
influence of Zs on fault component distance protection in cases where a PV power station is
• Previous studies did not analyze the adaptability of fault component distance protection on
connected to a power grid.
transmission lines connected to PV power stations. This study is the first to successfully perform
• Previous studies did not
the abovementioned analyze the adaptability of fault component distance protection on
analysis.
• transmission lines connected
The research ideas used in thisto study
PV power stations.
are also Thistostudy
applicable otheris the first
typical to protection
relay successfully perform
schemes
theonabovementioned analysis.
transmission lines connected to PV power stations.
• The research ideas used in this study are also applicable to other typical relay protection schemes
6. on
Conclusions
transmission lines connected to PV power stations.
This study derived the analytical expressions of fault currents on M side under different control
6. Conclusions
targets. Based on the fault-current characteristics, the adaptability of fault component distance
protection
This studywasderived
analyzed.
theFirst, the fault-current
analytical expressionscharacteristics on theon
of fault currents MMside differ
side significantly
under differentfrom
control
those of
targets. a synchronous
Based source and are
on the fault-current affected by a wide
characteristics, therange of factors,ofsuch
adaptability as component
fault the control targets,
distance
active and
protection was reactive power
analyzed. commands,
First, and voltage
the fault-current unbalance on
characteristics onthe
thepower
M side grid. Then,
differ the analysis
significantly from
showed that the equivalent impedance of M side tends to be large, and the phase
those of a synchronous source and are affected by a wide range of factors, such as the control angle of equivalent
targets,
impedance
active is greatly
and reactive affected
power by the active
commands, and reactive
and voltage power on
unbalance commands,
the power control
grid.targets, andanalysis
Then, the fault
conditions. In particular, when −180° < arg(Zs) < 0°, the fault component distance protection will
showed that the equivalent impedance of M side tends to be large, and the phase angle of equivalent
operate incorrectly.
impedance is greatly affected by the active and reactive power commands, control targets, and fault
Third, the analysis showed that the measured impedance is affected by the related parameters
conditions. In particular, when −180◦ < arg(Zs ) < 0◦ , the fault component distance protection will
in the control strategy, fault resistance, fault location, and fault type, among others. Unlike
operate incorrectly.
conventional synchronous generators, the measured impedance differs significantly in terms of
Energies 2019, 12, 1578 17 of 19

Third, the analysis showed that the measured impedance is affected by the related parameters in
the control strategy, fault resistance, fault location, and fault type, among others. Unlike conventional
synchronous generators, the measured impedance differs significantly in terms of magnitude and phase
angle, which will make the distance relay unable to calculate the actual fault impedance accurately.
Therefore, errors in impedance measurements can exacerbate problems with the adaptability of fault
component distance protection when PV power stations are interconnected with a power grid.
Finally, theoretical analysis and simulations showed that existing fault component distance
protection are not suitable for fault protection in transmission lines connected to PV power stations. If
a ground metallic fault occurs in the transmission lines, the fault component distance protection will
operate correctly. However, if the fault type is a phase-to-phase metallic fault or the fault with fault
resistance, the fault component distance protection may fail to operate.

Author Contributions: Investigation, Y.L.; methodology, Y.L. and G.X.; supervision, C.W.; validation, G.X.;
writing—original draft, G.X.; writing—review and editing, Y.L., W.Z., and C.W.
Funding: This study was supported by the National Natural Science Foundation of China (Grant No.51707193)
and the Fundamental Research Funds for the Central Universities (2017QJ05).
Conflicts of Interest: The authors declare no conflict of interest.

Appendix A

s
 .  √
∆UMϕϕ  = 3U+ 1 + kρ2 + 1/kλ2 + 2kρ cos(δ+ − δ− + θϕϕ )−
, (A1)
cos(∆δ) − 2kρ /kλ cos(δ+ − δ− − ∆δ + θϕϕ )
v
u
u  2
U+ cos δ+ + U− cos δ− + U0 cos δ0 − Us cos(δ+ − ∆δ)
 .  t
∆UMa  =  2 , (A2)
+ U+ sin δ+ + U− sin δ− + U0 sin δ0 − Us sin(δ+ − ∆δ)

−1 Us sin(δ − ∆δ) − U sin δ − U sin δ − U sin δ0

+ + + − − 0
!
∆θua = tan , (A3)
Us cos(δ+ − ∆δ) − U+ cos δ+ − U− cos δ− − U0 cos δ0

Nomenclature
U voltage magnitude
η initial phase angle of voltage
δ symbol standing for phase A, B, or C
k0 grounding coefficient
P0 power output of the PV power
Pref station during normal operation
Qref active power commands
Ivm reactive power commands
xd magnitude of the positive-sequence
ω current component
Zk measured impedance
∆Z actual fault impedance
Zset additional impedance
Rf setting impedance
kχ fault resistance
kρ control targets
kl grid voltage unbalance coefficient
kλ zero-sequence compensation
∆δ coefficient
Energies 2019, 12, 1578 18 of 19

References
1. Kouro, S.; Leon, J.I.; Vinnikov, D.; Franquelo, L.G. Grid-connected photovoltaic systems: An overview of
recent research and emerging PV converter technology. IEEE Ind. Electr. Mag. 2015, 9, 47–61. [CrossRef]
2. Jung, D.Y.; Ji, Y.H.; Park, S.H.; Jung, Y.-C.; Won, Y.-C. Interleaved soft-switching boost converter for
photovoltaic power-generation system. IEEE Trans. Power Electr. 2011, 26, 1137–1145. [CrossRef]
3. Karanayil, B.; Ceballos, S.; Pou, J. Maximum power point controller for large scale photovoltaic power plants
using central inverters under partial shading conditions. IEEE Trans. Power Electr. 2018, 99, 3098–3109.
[CrossRef]
4. Hong-Seok, S.; Kwanghee, N. Dual current control scheme for PWM converter under unbalanced input
voltage conditions. IEEE Trans. Ind. Electron. 1999, 46, 953–959. [CrossRef]
5. Baran, M.E.; El-Markaby, I. Fault analysis on distribution feeders with distributed generators. IEEE Trans.
Power Syst. 2005, 20, 1757–1764. [CrossRef]
6. Kabiri, R.; Donald, G.H.; Peter McGrath, B. Control of active and reactive power ripple to mitigate unbalanced
grid voltages. IEEE Trans. Ind. Appl. 2016, 52, 1660–1668. [CrossRef]
7. Nagpal, M.; Henville, C. Impact of power-electronic sources on transmission line ground fault protection.
IEEE Trans. Power Deliv. 2018, 33, 62–70. [CrossRef]
8. Chen, S.; Tai, N.; Fan, C.; Liu, J.; Hong, S. Adaptive distance protection for grounded fault of lines connected
with doubly-fed induction generators. IET Gener. Transm. Distrib. 2017, 11, 1513–1520. [CrossRef]
9. Bollen, M.; Hassan, F. Protection. In Integration of Distributed Generation in the Power System, 1st ed.;
Hooshyar, A., Azzouz, M.A., El-Saadany, E.F., Eds.; Wiley: Hoboken, NJ, USA, 2011; pp. 299–366.
10. Zhu, S.; Xing, Y.; Sui, F. Fault component reactance relay. IEEE Trans. Power Deliv. 1996, 11, 1292–1300.
11. Qing, L.; Zengping, W.; Yan, X. Study on the influence of TCSC on distance-relay fault protection. In
Proceedings of the Transmission and Distribution Conference and Exhibition: Asia and Pacific IEEE/PES,
Dalian, China, 18 August 2005; pp. 1–4.
12. Nengling, T.; Chen, C. A new weak fault component reactance distance relay based on voltage amplitude
comparison. IEEE Trans. Power Deliv. 2008, 23, 87–93. [CrossRef]
13. Khederzadeh, M. Performance evaluation of fault component distance relay for protection of lines
compensated by UPFC. In Proceedings of the 10th IET International Conference Developments in Power
System Protection (DPSP 2010), Managing the Change, Manchester, UK, 29 March–1 April 2010.
14. Yan, K.; Zhang, B.; Zhang, W.; Huang, R.; Jin, X. Analysis and improvement of relay protection for photovoltaic
power station connected to grid. In Proceedings of the 2013 IEEE International Conference of IEEE Region 10
(TENCON 2013), Xi’an, China, 22 October 2013; pp. 1–4.
15. Jia, K.; Gu, C.; Xuan, Z.; Li, L.; Lin, Y. Fault characteristics analysis and line protection design within a
large-scale photovoltaic power plant. IEEE. Trans. Smart Grid 2018, 9, 4099–4108. [CrossRef]
16. Li, B.; Yuan, Y.; Wei, L. Research on relay protection of grid-connected photovoltaic power station in the
plateau. In Proceedings of the 2011 International Conference on Advanced Power System Automation and
Protection, Beijing, China, 16–20 October 2011; pp. 1814–1818.
17. Xue, S.; Yang, J.; Chen, Y.; Wang, C.; Shi, Z.; Cui, M.; Li, B. The applicability of traditional protection methods
to lines emanating from VSC-HVDC interconnectors and a novel protection principle. Energies 2016, 9, 400.
[CrossRef]
18. Alam, M.M.; Leite, H.; Liang, J.; Carvalho, A.-D.S. Effects of VSC based HVDC system on distance protection
of transmission lines. Int. J. Electr. Power Energy Syst. 2017, 92, 245–260. [CrossRef]
19. Liu, Y.; Li, G.; Wang, H.; Wang, Z.; Yu, M. Research on AC line distance relay in the presence of modular
multilevel converter based HVDC. In Proceedings of the 2016 IEEE PES Asia-Pacific Power and Energy
Engineering Conference (APPEEC), Xi’an, China, 25–28 October 2016; pp. 1622–1626.
20. Hooshyar, A.; Azzouz, M.A.; El-Saadany, E.F. Distance protection of lines emanating from full-scale
converter-interfaced renewable energy power plants—Part I: Problem statement. IEEE Trans. Power Deliv.
2015, 30, 1770–1780. [CrossRef]
21. Hooshyar, A.; Azzouz, M.A.; El-Saadany, E.F. Distance protection of lines emanating from full-scale
converter-interfaced renewable energy power plants—Part II: Solution description and evaluation. IEEE
Trans. Power Deliv. 2015, 30, 1781–1791. [CrossRef]
Energies 2019, 12, 1578 19 of 19

22. Jia, J.; Yang, G.; Nielsen, A.H.; Ronne-Hansen, P. Impact of VSC control strategies and incorporation of
synchronous condensers on distance protection under unbalanced faults. IEEE Trans. Ind. Electr. 2019, 66,
1108–1118. [CrossRef]
23. Ma, J.; Zhang, W.; Liu, J.; Thorp, J.S. A novel adaptive distance protection scheme for DFIG wind farm
collector lines. Int. J. Electr. Power Energy Syst. 2018, 94, 234–244. [CrossRef]
24. He, L.; Liu, C.C.; Pitto, A.; Cirio, D. Distance protection of AC grid with HVDC-connected offshore wind
generators. IEEE Trans. Power Deliv. 2014, 29, 493–501. [CrossRef]
25. Tang, J.; Song, G.; Wang, X.; Wang, C. A novel directional relay applicable to power system with wind farms.
In Proceedings of the 2016 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC),
Xi’an, China, 25–28 October 2016; pp. 2117–2121.
26. Chen, Y.; Wen, M.; Yin, X.; Cai, Y.; Zheng, J. Distance protection for transmission lines of DFIG-based wind
power integration system. Int. J. Electr. Power Energy Syst. 2018, 100, 438–448. [CrossRef]
27. Yazdani, A.; Iravani, R. A unified dynamic model and control for the voltage-sourced converter under
unbalanced grid conditions. IEEE Trans. Power Deliv. 2006, 21, 1620–1629. [CrossRef]
28. Fang, Y.; Jia, K.; Yang, Z.; Bi, T.; Li, Y. Impact of inverter-interfaced renewable energy generators on distance
protection and an improved scheme. IEEE Trans. Ind. Electron. 2018. [CrossRef]
29. Jia, K.; Li, Y.; Fang, Y.; Zheng, L.; Bi, T.; Yang, Q. Transient current similarity based protection for wind farm
transmission lines. Appl. Energy 2018, 225, 42–51. [CrossRef]
30. Bergen, A.R.; Vittal, V. Generator Modeling II (Circuit Viewpoint). Power Systems Analysis, 2nd ed.; Prentice
Hall: New Jersey, NJ, USA, 1999; pp. 215–269.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).