2019 IEEE Fourth Ecuador Technical Chapters Meeting (ETCM)

Determination of Overcurrent Protection Settings

Based on Estimation of Short-Circuit Currents
Using Local Measurements
Fernando Chapi Antonio Fonseca Fabián Pérez
CELEC EP - Transelectric CELEC EP - Transelectric Escuela Politécnia Nacional
Escuela Politécnica Nacional Escuela Politécnica Nacional Facultad de Ingeniería en Eléctrica y
Facultad de Ingeniería Eléctrica y Facultad de Ingeniería Eléctrica y Electrónica
Electrónica Electrónica Quito, Ecuador
Quito, Ecuador Quito, Ecuador [email protected]
[email protected] [email protected]

Abstract—On Electrical Distribution Systems (EDSs), ensure proper operation of overcurrent protection when there
system changes upstream of overcurrent protective devices are topological changes due to a transmission system
result in variations of the Thevenin equivalent and operation condition [5]. Icc levels are continuously changing
consequently in variations of short-circuit currents (Icc). in the EDSs due to changes in topology and therefore due to
Changes in the Thevenin equivalent are caused by some changes in the grid impedance [6]. A variation on the Icc
operating condition or by growth of transmission system. In levels while keeping overcurrent protection settings fixed can
that condition, the fixed settings in overcurrent protection produce a miscoordination, causing loss of selectivity. A
could cause selectivity problems in the protection systems. This feasible solution to this type of problems is to use adaptive
paper proposes the development of an overcurrent protection
overcurrent protections.
scheme. The main idea is to change the settings based on the
Icc of EDS. By adaptivelychanging settings allows mitigation of Adaptive protection is defined as the ability to modify the
possible miscoordinations caused by Icc variation effects. The response of protection against changes in system conditions
IEC 60909 standard is used to calculate the Icc, by using a or requirements, generally performed in real time
prefault Thevenin equivalent, wich is obtained by the voltage automatically or by human intervention [7], [8], [9]. There
and current measurements acquired at the relay location. The are two alternatives for modifying settings on adaptive
proposed methodology is developed on PowerFactory - relaying. The first implements the adaptive logic at the relay
DIgSILENT software and applied to the IEEE Distribution 13- level, so the relay modifies its parameters based on the input
Bus Test Feeder System.
signals (voltage, current, binary inputs). The second
Keywords—IEC 60909 Standard, Overcurrent Protection, alternative uses a communication link to remotely adjust the
Thevenin Equivalent settings, by allowing the adaptive function to be performed at
the substation level or at the system level [10], [11].
I. INTRODUCTION
For instance, the authors in [5] proposed a variable
Electrical Distribution Systems (EDSs) are exposed to setting of overcurrent protection during the connection or
diverse abnormal operating conditions that cause a change in disconnection of distributed generation, where the tap or
their electrical variables. These abnormal conditions are pick-up is determined based on load variations and the dial is
known as disturbances, with short-circuits being the most determined to have similar operating time to the one set.
common and frequent that exist in power systems, usually
causing very high currents. When short-circuits do not clear Another publication about adaptive overcurrent
in time, they can cause severe damage or destruction of protection is presented in [6] where an overcurrent relay
power system components. To avoid these damages, the coordination algorithm is proposed using an adaptive diffuse
EDSs are provided with protective devices that isolate the technique to select the optimal relay settings.
elements in faulted condition [1], [2]. On the other hand, the Most of the research on adaptive overcurrent protection
electricity that is generated in power plants is supplied to the have been only focused its efforts on inverse time phase
EDS through the transmission systems to which it is overcurrent protection considering changes in the network
connected. Grid topology and generation dispatch has a huge topology or due to the connection of distributed generation.
impact on the determination of the short-circuit current levels In this work, it is proposed to develop a scheme of
that exist within the EDS. Consequently, changes that occur overcurrent protection (phase and ground) with the ability to
in the generation and transmission systems upstream of the vary the settings depending on the Icc that is presented in the
EDS must be considered in the protection system settings. EDS. The Thevenin equivalent is used to determine the Icc
The overcurrent protection is the most used in the EDSs obtained as a function of the measurement of voltages and
due to its low cost, easy installation and fast configuration. currents acquired at the relay location, also known as local
However, this type of protections is the most likely to have a measurements.
miscoordination since its performance depends directly on For this, a simulated overcurrent relay is developed in
the short-circuit current levels (Icc) [3], [4]. Currently, the PowerFactory DIgSILENT Simulation Language (DSL) and
overcurrent protection settings (tap, dial and curve type) are implemented in PowerFactory DIgSILENT Programming
defined considering a coordination from the load towards the Language (DPL) making it possible to develop the proposed
source, and they are determined by a predominant methodology. Furthermore, the methodology is verified with
topological condition. Once set, they remain fixed for all simulation of the IEEE Distribution 13-Bus Test Feeder
other operating conditions. In addition, it is difficult to [12].

978-1-7281-3764-3/19/$31.00 ©2019 IEEE

Authorized licensed use limited to: University of Exeter. Downloaded on May 06,2020 at 11:24:06 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE Fourth Ecuador Technical Chapters Meeting (ETCM)

II. THEORETICAL FRAMEWORK AND METHODOLOGY If the voltage and current measurement point (local
To determine the tap and dial of the overcurrent measurement) is found in bus k, from (2) the known
protection based on the short-circuit current, the present variables are g, h, u and w, and the unknowns that represent
work has been divided into three parts or three calculation the Thevenin equivalent are R , X , E , E [13].
blocks. To solve (2), least squares method is used, which uses
In Fig. 1, the first part calculates the Thevenin impedance two measurements in bus k, considering that the Thevenin
using local measurements (voltage and current equivalent remains constant in the time interval in which the
measurements), in the second part, short-circuit currents are two measurements are taken and the system is in a steady
determined according to IEC 60909 from the Thevenin state.
impedance and primary and backup protection zones. In the In the proposed methodology, to correctly determine the
third part the overcurrent protection settings are defined Thevenin sequence impedances, an unbalanced system is
according to the short-circuit current. considered, which when solving (2) using positive, negative
and zero sequence measurements of voltage and current
(local measurements), positive, negative and zero sequence
impedances are obtained respectively.
B. Calculation of Short Circuit Currents Using IEC 60909
Standard and prefault calculated Thevenin Impedance
[14], [15]
Continuing with the general diagram (Fig. 1), in this
second part, the calculation of the short-circuit currents using
the Thevenin sequence impedances and the sequence
impedances of the primary and backup protection zones is
Fig. 1. General diagram of adaptive overcurrent protection.
described.
A. Thevenin Prefault Impedance Calculation Using Local It is considered that the simulated faults are far from the
Measurements. generator since they are calculated in the feeders of the
The electrical system shown in Fig. 2 is considered to distribution system, therefore the initial symmetrical short-
determine the prefault Thevenin impedance. circuit current I ´´, and the permanent short-circuit current I
are equal [16], [17].
The procedure for calculating short-circuit currents
according to IEC 60909, consists in reducing the system by
an equivalent short-circuit impedance Z at the faulted point.
In this case the equivalent impedance at the faulted point is
Fig. 2. Two-bus electric system the sum of the Thevenin impedance calculated using local
The following equation is valid for the electrical system measurements and the impedance of the primary or backup
of Fig. 2. protection zone.

E =V +Z ∗I (1) C. Overcurrent Protection Settings in function of Short

Circuit Current
where: The proposed protection scheme considers the use of
E it is the voltage at the source: = + ∗ adaptive relays at the feeders’ sending end, as seen in Fig. 3,
relay C2 has the characteristic of varying its settings
V it is the voltage in the load: V = u + j ∗ w (adaptive relay) and relay C1 has fixed settings using Non-
directional overcurrent function as backup protection.
I it is the system current: I = g + j ∗ h
To determine the overcurrent protection settings based on
Z is the Thevenin impedance: Z = R + j ∗ X the short-circuit current (adaptive protection), the use of two
Expressing (1) in matrix form and considering protection units, one with instantaneous overcurrent as
rectangular coordinates yield: protection of the primary zone (generally 80% of the
E protected feeder), and a reverse time overcurrent protection
1 0 −g h E u for the backup zone (backup can cover faults in the entire
∗ = (2) feeder of a EDS).
0 1 −h −g R w
X
Upstream System from the adaptative relay Backup protection zone
Primary protection zone Icc_min
T Icc_max
CT CT
52 52 LsT LsT
PT Load
I
C1 C2 v

51/51N 50/50N
51/51N
Fig. 3. EDS proposed using an adaptive relay C2 in the header of the feeder.

Authorized licensed use limited to: University of Exeter. Downloaded on May 06,2020 at 11:24:06 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE Fourth Ecuador Technical Chapters Meeting (ETCM)

For adaptive inverse time protection, it is considered that Finally, the dial setting varies depending on the
the adaptive relay C2 has as coordination limit the maximum short-circuit current, the 51/51N tap defined above
overcurrent curves of the upstream protective device (relay and the upstream overcurrent protection settings.
C1 located behind the adaptive relay). That is, relay C2
needs to coordinate with the settings set in the upstream The following equation can be written considering as a
overcurrent protection, so tap and appropriate dial are selectivity criterion, to maintain a minimum coordination
determined that allows to maintain an adequate coordination time interval ∆t1between the timed overcurrent relays.
time interval (∆t) between inverse-time charactetistics during t = t + ∆t1 (3)
the most critical fault. t , t : Operating times of relays C1 and C2 respectively.
Therefore, adaptive protection requires more setting ∆t1 : Coordination criteria – setting of the adaptive
requirements that can be grouped into: data of the primary protection.
zone, backup zone, overcurrent protection settings for
coordination (upstream protection settings) and coordination Replacing the time overcurrent characteristic in (3) and
criteria. finding the relay C2 dial gives:

1) Setting the instantaneous overcurrent protection unit _

Variations of the Thevenin equivalent cause the reach of ∗ ∆ ∗
_
the instantaneous overcurrent protection unit (50) to change.
Namely, it can overreach or underreach depending on the = _ (4)
value of the fault current contribution to the short circuit. _

To reduce overreach or underreach, with the adaptive Equation (4) determines the relay C2 dial, knowing the
instantaneous overcurrent protection unit, it is proposed to setting parameters ( I _ , dial , k , α_c1 , C ) of the
use a variable tap setting that allows to have a constant upstream relay C1 and maximum short-circuit current used
protection reach. to set the instantaneous protection unit.
The 50/50N tap of relay C2 in Fig. 3 is set with the Before calculating dial , the range of selectivity
maximum short-circuit current (Icc_max), calculated between curve C1 and C2 is evaluated considering the
according to the Thevenin impedance (determined with local minimum current in which coordination must exist, i.e. in the
measurements) and the impedance of the protected line. To current equal to the tap of curve C1.
obtain the Icc_max, using the IEC 60909 standard, all types Therefore, using the tap current of curve C1, the
of short-circuits are calculated at a percentage ( %) of the coordination time between the two curves can be evaluated
protected line, where is generally 80% [18]. using (5).
Therefore, to adjust the instantaneous phase overcurrent
protection unit (50) and residual (51N), the maximum short- ∆ 2≥ ∗ _ + − ∗ _ + (5)
circuit current existing in the phases and in the ground (3Io) _ _
respectively is used. If the coordination time interval (∆ 2) in the tap current of
2) Setting the inverse time overcurrent protection unit C1, is greater than or equal to the selectivity interval input to
First of all, it is considered that there is no variation in the relay C2, dial calculated above is considered as the
type of curve, which has to be properly defined in the initial appropriate setting, otherwise dial is recalculated with (4)
protection coordination study. When adjusting the tap or using the tap current of curve C1 as a reference.
pick-up for 51/51N of relay C2 in Fig. 3, it is given the The definition of the settings for the ground fault
ability to vary depending on the maximum load current, the overcurrent protection is similar to the one stated above, and
minimum short-circuit current (Icc_min) and the tap of the differs only in the load currents (I_load) and short-circuit
upstream overcurrent protection. Where the Icc_min is currents entering the block, which are the currents (3Io)
calculated based on the Thevenin impedance and the circulating through the ground of the relay.
impedance of the backup zone.
III. DIGITAL ALGORITHM IMPLEMENTATION
In this case, the tap or pick-up of relay C2 in Fig. 3, is
defined as the minimum short-circuit current as long as the Figure 4 shows the flow chart elaborated by
value of the Icc_min is within the range of maximum load programming in PowerFactory - DIgSILENT DPL, which
current and the tap of the upstream overcurrent protection. interacts with the DSL blocks to obtain or transmit the input
or output variables respectively and develop the proposed
However, if the Icc_min is greater than the tap set in the methodology.
upstream overcurrent protection (relay C1 in Fig. 3), the tap
or pick-up of the adaptive protection takes the value of the IV. SIMULATION AND RESULTS
tap of relay C1 as a new setting. This is done because by not To present the results of the proposed methodology, the
fulfilling this condition, there may be miscoordination for IEEE Distribution 13-Bus Test Feeder is used [12]; because
faults with currents similar to the tap of the upstream the test system does not show the details of the equivalent of
overcurrent protection. Otherwise, if the Icc_min is less than the system located at bus 650, a transformer (T) is used to
the maximum load current, the adjusted tap is equal to the couple the distribution network with the subtransmission
maximum load current with a safety factor ( ), in this case system (Fig. 5).
there is an alarm that indicates a reduction of the backup
zone.

Authorized licensed use limited to: University of Exeter. Downloaded on May 06,2020 at 11:24:06 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE Fourth Ecuador Technical Chapters Meeting (ETCM)

Start The variation of the load, considers that the power flow
of the feeder has a residential tendency and the
Power Flow
measurements for this application are made every 30
Access to the first
measurement
minutes, that is, in this study case there are 48 iterations.
Run the event in
Each iteration is a stationary state of the simulated system, in
Load variation
iteration i which there is an unbalanced power flow solution.
Measurement Power Flow
Is there an event
update
Access to the
To simulate the variation of the Thevenin equivalent, the
-First measurement
in iteration i?
equal to the second second transmission lines 4, 3, 2, are sequentially operated, in
measurement
measurement iterations 12, 24 and 36, respectively.
Thevenin Impedance
(Zth) Calculation
In the initial state (the Thevenin equivalent starts with
Short Circuit
Storage of Currents Calculation
Information input: four L/T in operation), the settings of the two relays (R1, R2)
-Primary protection zone
information for using Zth
-Backup protection zone are defined, considering conventional relays, i.e. they have
analysis (results,
graphs) Determination of fixed settings.
i=i+1 operating times and Information input:
new protection
settings
- Overcurrent protection Considering relay R2 as an adaptive protection, the
settings for coordination
(upstream protection following is adjusted: the primary protection zone with 80%
¿Iteration i=n?
settings) and coordination
criteria
of the line L_650-632, the backup zone covers up to the line
L_671-680 so the settings are the sum of the impedances of
the lines L_650-632, L_632-671, and L_671-680, the
End
overcurrent protection settings for coordination are the fixed
Fig. 4. Flowchart to determine the overcurrent protection settings based on
the short-circuit current
settings of the relay R1 and a ∆t of 300 ms with =30% is
used as coordination criteria.
~
V
In order to observe the underreaching condition that the
~
V
SNT
primary zone can have while maintaining fixed settings, a
single-phase short-circuit is simulated at 75% of the
Z_SNI

Z_SNI

Z_SNI

Z_SNI
LT_1

LT_2

LT_3

LT_4

SNT
Z_SNI

Z_SNI

Z_SNI

Z_SNI
LT_1

LT_2

LT_3

LT_4

HV
L/T 1 L/T 4
L/T 1 L/T 4
protected line L_650-632 from iteration 42.
15/0.4 kV 400 kVA

T HV
T

15/0.4 kV 400 kVA

R1
T
R650

R2
T
LOHL650-632
LineConfig6..

Primary zone

646 645 632

Backup zone

633 634
R650
R1
Figure 6 shows the Thevenin impedance calculated by
R2
the adaptive relay R2 during the 48 iterations, observing the
XFM-1
XFM-1Typ

LOHL645-646 LOHL632-645 LOHL632-633

LineConfig6.. LineConfig6.. LineConfig6..
LOHL650-632
DistributedLoad

LineConfig6..
LineConfig601val
s4LOHL632-671

L646-DcZ L645-YcPQ L634-YcPQ

variation of the Thevenin impedance. The Thevenin zero

Primary zone
LodYEcPQ

LodPPcZ LodPEcPQ LodYEcPQ

Switch

Backup zone
sequence impedance calculated remains constant throughout
611 684 671 692 675

LC692-675
606val

LOHL684-611 LOHL671-684
LC684-652A

LineConfig6.. LineConfig6..
607 Aval

the simulation, because the type of transformer connection

LOHL671-680
LineConfig6..

632 633
C611 C675
L611-YcI L571-DcPQ L692-DcI L675-YcPQ
LodPEcI LodDcPQ LodPPcI LodYEcPQ

652

L652-YcZ
LodPEcZ
680

OHL632 645
located at node 650 is Dyn5.
Thevenin impedance [Ω]
Fig. 5. Single-line diagram of the simulated 13-node IEEE test feeder in 0,3961
X1 = 12,000 X1 = 24,000 X1 = 36,000
PowerFactory - DigSilent. [-] ---- Rth(1,2) [Ω] vs Iteration # iteration = 48.0
The Thevenin equivalent is represented with an ideal 0,3165 ---- Xth(1,2) [Ω] vs Iteration Xth_(1,2) = 0.378 ohms
voltage source and four transmission lines (L/T) in parallel. ---- Rth(0) [Ω] vs Iteration
In the simulation, the operation (connection/disconnection) 0,2368 ---- Xth(0) [Ω] vs Iteration
# iteration = 48.0
of the transmission lines represents the variation of the 0,1572
Rth_(1,2) = 0.065 ohms
Thevenin impedance in bus 650. # iteration = 22.0 # iteration = 45.0
Rth_(0) = 0.016 ohms
Xth_(0) = 0.082 ohms
0,0775
In addition, two relays, R1 and R2 are modeled, located
on the low side of the transformer and in the main feeder -0,0021
Iteration
position respectively. Relay R2 protects line L_650-632 0,00 10,00 20,00 30,00

Fig. 6. Case 1.- Thevenin impedance calculated by the R2 relay

40,00 [-] 50,00

against failures and relay R1 is considered as a backup

protection of the transformer. When there is a variation of the Thevenin impedance,
there is also a variation of the overcurrent protection settings
To depict the functionality of the adaptive protection, a
as shown in Fig. 7, the red line represents the setting of the
comparison of the operating times between the use of
adaptive tap on the 50/50N function of the R2 protection, the
conventional relays (fixed settings) and the proposed scheme
green line represents the fixed setting of the instantaneous
(variable settings) is made, considering two case studies that
overcurrent function, and the blue line is the phase current
present the tap, dial and the Thevenin equivalent graphically.
(Fig. 7-a) and ground (Fig. 7-b) measured by the relay in
Variations of these cases are developed in [19].
each iteration for protection 50 and 50N respectively.
Case 1.- Reduction of the Icc through the increase of the
In Fig. 7, from iteration 42 the current measured by the
Thevenin impedance, where the instantaneous adaptive
relay is increased to the fault value, in this case, by the type
function is analyzed.
of short circuit, the current flowing through phase A is equal
Case 2.- Increase of the Icc through the reduction of the to the residual current or circulating current through the
Thevenin impedance, where the adaptive timed function is ground (3I0).
analyzed.
Also, from Fig. 7 it is observed that, for the type of
A. Case 1.- Increase of the Thevenin Impedance. simulated fault, the 50N function is the first to operate, since
In this case, a day of operation of the EDS is simulated, the Icc (4501 primary Amps) exceeds the adjustment
in which simultaneously there are variations of the Thevenin threshold (4354 primary Amps) defined by the adaptive
equivalent and changes in the feeder load. protection R2. In addition, having a fixed setting (green line),

Authorized licensed use limited to: University of Exeter. Downloaded on May 06,2020 at 11:24:06 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE Fourth Ecuador Technical Chapters Meeting (ETCM)
Thevenin impedance [Ω]
the 50N protection functionality is lost, since the current 3I0 0,8665
X1 = 12,000 X1 = 24,000 X1 = 36,000

of the single-phase short circuit is less than the initially set [-]
# iteration = 2.0 ---- Rth(1,2) [Ω] vs Iteration
operating current, that is, there is an underreach of the 50N 0,6885 Xth_0 = 0.826 ohms ---- Xth(1,2) [Ω] vs Iteration
protection. # iteration = 1.0 ---- Rth(0) [Ω] vs Iteration
Xth_(1,2) = 0.378 ohms
Settings / Phase current [A]
0,5105
---- Xth(0) [Ω] vs Iteration
8127,5
X1 = 12,000 X1 = 24,000 X1 = 36,000
0,3325
[-] Y1 =7762,000
Rth_0 = 0.057 ohms
6467,9
# iteration = 16.0 0,1545
Rth_(1,2) = 0.033 ohms
50_Pick-up = 7230.4 A-prim
4808,4 -0,0235
# iteration = 37.0 0,00 10,00 20,00 30,00 40,00 [-] 50,00
Iteration
50_Pick-up = 4667.3 A-prim
Fig. 8. Case 2 - Thevenin impedance calculated by relay R2.
3148,9
# iteration = 45.0
Icc_1F = 4501.8 A-prim When the three-phase short circuit (100% of the L_650-
# iteration = 27.0
1489,3 I_Load = 500.3 A-prim 632 line) occurs at iteration 42, the current measured by the
relay (blue line) increases to 6,842 primary Amps, and as
-170,22
0,00 10,00 20,00 30,00 40,00 [-] 50,00
Iteration seen in Fig. 9, the Icc exceeds the fixed setting, but does not
---- Pick-up 50 adaptive [A] vs Iteration ---- Pick-up 50 fixed [A] vs Iteration exceed the adaptive setting defined by the R2 relay.
---- Relay R2 phase current [A] vs Iteration Settings / Phase current [A]
Settings / 3Io [A] a) 50 function 9127,4
X1 = 12,000 X1 = 24,000 X1 = 36,000
X1 = 12,000 X1 = 24,000 X1 = 36,000 # iteration = 40.0
6471,5 [-] 50_Pick-up = 7750 A-prim
Y1 =5764,000 # iteration = 18.0
[-] 7268,3
50_Pick-up = 6356 A-prim
5129,2
# iteration = 47.0
# iteration = 16.0 5409,2 Icc_3F = 6842 A-prim
50N_Pick-up = 5585 A-prim
3786,9
# iteration = 38.0 Y1 =4664,000
50N_Pick-up = 4354 A-prim 3550,1

2444,6 # iteration = 44.0

Icc_1F = 4501.8 A-prim # iteration = 27.0
1691,0
I_Load = 500.3 A-prim
# iteration = 26.0
1102,3 3I0 = 109.5 A-prim
-168,12
0,00 10,00 20,00 30,00 40,00 [-] 50,00
Iteration
-240,05
0,00 10,00 20,00 30,00 40,00 [-]
50,00
---- Pick-up 50 adaptive [A] vs Iteration ---- Pick-up 50 fixed [A] vs Iteration
Iteration
---- Pick-up 50N adaptive [A] vs Iteration ---- Pick-up 50N fixed [A] vs Iteration ---- Relay R2 phase current [A] vs Iteration
---- Unbalance current (3Io) of relay R2 [A] vs Iteration Fig. 9. Case 2. - Variation of the 50 function setting of the R2 relay
b) 50N function
Fig. 7. Case 1.- Variation on the 50 function setting of the R2 relay In other words, for the simulated three-phase fault, the
protection with fixed settings operates incorrectly and the
B. Case 2.- Reduction of the Thevenin impedance protection with adaptive adjustments R2 operates correctly
To observe the variation of the zero sequence impedance maintaining the appropriate selectivity.
of the Thevenin equivalent, the transformer connection group To have correct adaptive protection functionality, when
(T) is changed to YNyn0 and the equivalent initially starts the instantaneous overcurrent protection (50) does not
with a transmission line (L/T) in operation. operate, the backup protection 51 must operate according to
In the initial state (one L/T) the settings of the two relays the settings defined by the adaptive relay R2.
(R1, R2) are defined considering conventional relays, that is, According to the proposed methodology, the adaptive
they have fixed settings. relay defines protection settings 51/51N in each iteration,
The total load of the feeder presented in Case 1 is used, which are detailed in Fig. 10.
i.e. the simulation has 48 iterations. To simulate the variation The tap of function 51 (red line in Fig. 10-a), before
of the Thevenin equivalent, the transmission lines 4, 3, 2, are entering into operation L/T 2 remains constant because the
sequentially operated, in iterations 12, 24 and 36, minimum short-circuit current (calculated according to the
respectively. Thevenin impedance and the backup protection zone) is
To observe the underreaching condition of the primary greater than the load current and less than the tap of the
protection zone by maintaining fixed settings and the upstream protection R1.
coordination of the inverse time overcurrent functions, a When the L/T 2, 3 and 4 enter into operation, the
three-phase short circuit at 100% of the protected line Thevenin impedance decreases causing the increase of the
L_650-632 is simulated at iteration 42. minimum short-circuit current to values lower than the load
Figure 8 shows the Thevenin impedance calculated by current, so at these intervals and according to the proposed
the adaptive relay R2 during the 48 iterations. Upon entering methodology, the tap of function 51 is defined by the
the L/T 2 in service, the Thevenin impedance decreases, maximum load current.
since the resulting Thevenin impedance is the sum of two In Fig. 11, the adaptive protection settings are plotted at
impedances, the transformer impedance (T) and the parallel the time of the three-phase short circuit at 100% of the
of the L/T's 1 and 2. protected line, the green line represents the fixed setting of
When a transformer (T) with the connection group the conventional protection R2 (defined with one L/T) and
YNyn0 is considered, the zero sequence impedance the blue line represent the adaptive protection settings from
variations in the system upstream of the adaptive relay R2 iteration 42 (settings when three-phase short circuit occurs).
are reflected in the Thevenin equivalent, as shown in Fig. 8.

Authorized licensed use limited to: University of Exeter. Downloaded on May 06,2020 at 11:24:06 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE Fourth Ecuador Technical Chapters Meeting (ETCM)
Settings / Phase current [A]
10000
X1 = 12,000 X1 = 24,000 X1 = 36,000
one hand, to solve the reach problems of the instantaneous
protection functions (50/50N), maintaining an almost
[-]
# iteration = 47.0
Icc_3F = 6842 A-prim
constant primary protection zone and, on the other hand, with
---- Pick-up 51 adaptive [A] vs Iteration the inverse time functions (51/51N), a greater coordination
1000
---- Relay R2 phase current [A] vs Iteration time between adaptive protection and downstream
protections, without neglecting the appropriate coordination
time interval (∆t).
# iteration = 48.0
Pick-up_51 = 462 A-prim In a radial system, without internal generation, the
# iteration = 4.0
100
I_Load = 236 A-prim proposed methodology determines the adaptive protection
0,00 10,00 20,00 30,00 40,00 [-] 50,00
Iteration settings based on the short-circuit currents evaluated with the
a) Pick-up of function 51
Settings
0,1440
X1 = 12,000 X1 = 24,000 X1 = 36,000 IEC 60909 standard, and these are defined based on the
impedances of the protection zones (primary zone and
[-]
---- Dial 51 adaptive [A] vs Iteration backup zone) and the Thevenin equivalent calculated with
0,1321

local measurements (voltage and current) at the location of

0,1202
the adaptive relay.

0,1084 # iteration = 47.0 VI. REFERENCES

Dial_51 = 0.100
[1] C. R. Mason, The art and science of protective relaying. Wiley, 1956.
0,0965
[2] L. Brand Contreras, Moncada V, and Juan A, Protecciones de
sistemas eléctricos. Santiago de Chile: Hugo E. González A., 1976.
0,0846
0,00 10,00 20,00 30,00 40,00 [-] 50,00 [3] M. Ojaghi, Z. Sudi, and J. Faiz, “Implementation of Full Adaptive
Iteration
b) Dial of function 51 Technique to Optimal Coordination of Overcurrent Relays,” IEEE
Fig. 10 Case 2 - Variation of the setting of function 51 of relay R2. Trans. Power Deliv., vol. 28, no. 1, pp. 235–244, Jan. 2013.
[4] P. Mahat, Z. Chen, B. Bak-Jensen, and C. L. Bak, “A Simple
Furthermore, in Fig. 11 the blue vertical line represents Adaptive Overcurrent Protection of Distribution Systems With
Distributed Generation,” IEEE Trans. Smart Grid, vol. 2, no. 3, pp.
the three-phase Icc (6,842 primary Amps) at 100% of the 428–437, Sep. 2011.
protected line, and the crossing points with the coordination
[5] S. Mitra, G. Misra, and P. Chattopadhyay, “A Smart Approach for
curves are the operating times of the protective devices. Development of an Adaptive Overcurrent Relay,” in 2018 Fifth
t [s] R1 International Conference on Emerging Applications of Information
IEC_Normal_Inverse
Ipset: 902,00 sec.A Technology (EAIT), 2018, pp. 1–4.
Tpset: 0,17
R2_Conventional [6] M. Singh and V. Telukunta, “Adaptive over current relay
IEC_Normal_Inverse
Ipset: 902,00 sec.A coordination algorithm for changing short circuit fault levels,” in
Tpset: 0,10
2015 IEEE Innovative Smart Grid Technologies - Asia (ISGT ASIA),
R2_Adaptive
1 IEC_Normal_Inverse 2015.
Ipset: 462,00 sec.A
0.585 s Tpset: 0,10
[7] G. D. Rockefeller, C. L. Wagner, J. R. Linders, K. L. Hicks, and D. T.
R2_Conventional
Instantaneous Rizy, “Adaptive transmission relaying concepts for improved
Ipset: 4664,00 sec.A
0.263 s Tpset: 0,05 performance,” IEEE Trans. Power Deliv., vol. 3, no. 4, pp. 1446–
R2_Adaptive 1458, Oct. 1988.
Instantaneous
0,1
Ipset: 7750,00 sec.A
Tpset: 0,05 [8] J. A. Kumar, S. S. Venkata, and M. J. Damborg, “Adaptive
0.060 s
transmission protection: concepts and computational issues,” IEEE
Trans. Power Deliv., vol. 4, no. 1, pp. 177–185, Jan. 1989.
I =6842,000 pri.A

50/51 - Phase Overcurrent

[9] A. C. Enríquez, “Protección adaptiva de sobrecorriente,” Universidad
Autónoma de Nuevo León, 1996.
I [A] [10] S. J. Jiménez Martínez, “Protección adaptable de sobrecorriente
instantánea.,” masters, Universidad Autónoma de Nuevo León, 2017.
0,01
6 kV 100 1000 10000 100000 1000000 [pri.A] 1000000

Fig. 11. Case 2. - Settings 51/50 of adaptive relay R2, from iteration 42
[11] F. C. Souza and B. A. Souza, “Adaptive overcurrent adjustment
For the simulation of case 2, the conventional relay settings: A case study using RTDS®,” in 2013 IEEE PES Conference
operates instantaneously producing miscoordination with on Innovative Smart Grid Technologies (ISGT Latin America), 2013,
pp. 1–5.
downstream relays; however, the use of adaptive settings
[12] “Distribution Test Feeders - Distribution Test Feeder Working Group
causes the relay R2 to operate in a coordinated manner, - IEEE PES Distribution System Analysis Subcommittee.”
providing adequate time for downstream protections to [13] H. Yuan and F. Li, “A comparative study of measurement-based
operate before relay R2 operates. It should be noted that, Thevenin equivalents identification methods,” in 2014 North
with the settings defined by the adaptive relay R2, there is an American Power Symposium (NAPS), 2014, pp. 1–6.
adequate coordination time interval ∆t = 322 ms in the [14] IEC Short-Circuit Currents in Three-Phase A.C. Systems - Part 0:
operating time of the relay R1 and the adaptive relay R2. Calculation of Currents, IEC 60909-0, 2001.
[15] “Recommended Practice for Calculating AC Short-Circuit Currents in
V. CONCLUSIONS Industrial and Commercial Power Systems,” IEEE Std 551-2006
Violet Book, pp. 1–308, Nov. 2006.
Operational variations or changes in the topology of the
[16] Nasser D. Tleis, Power Systems Modelling and Fault Analysis Theory
system upstream of the relay location, produce changes in and Practice.
short-circuit current levels causing overreach or underreach [17] Kasikci, Short Circuits in Power Systems: A Practical Guide to IEC
in conventional overcurrent relays that have fixed settings, 60 909. Wiley-VCH, 2002.
however, the use of adaptive protection reduces these [18] “Criterios de ajuste y coordinación de los sistemas de protección del
problems. SEIN.” Comité de Operación Económica del Sistema Interconectado
Nacional (COES SINAC).
The developed methodology gives the overcurrent relay [19] F. Chapi, “Determinación de ajustes de la protección de
the advantage of varying its settings depending on the short- sobrecorriente en base a la estimación de corrientes de cortocircuito
circuit current evaluated at the relay location. This allows, on usando mediciones locales,” Escuela Politécnica Nacional, 2019.

Authorized licensed use limited to: University of Exeter. Downloaded on May 06,2020 at 11:24:06 UTC from IEEE Xplore. Restrictions apply.