Testing Impedance Characteristics
Monir Hossain, PhD Milton I. Quinteros, PhD Hussain Al Marhoon, PhD
Relay Settings & Configurations Transmission Design Basis Transmission Protection
Entergy Services Inc. Entergy Services Inc. Entergy Services Inc.
639 Loyola Ave, New Orleans, 639 Loyola Ave, New Orleans, 639 Loyola Ave, New Orleans,
LA-70113 LA-70113 LA-70113
[email protected] [email protected] [email protected] Abstract – This paper presents a methodology to test Mho characteristics by applying AC phasor quantities at the inputs
and Quadrilateral impedance characteristics. The methodology of the protective relay and sensing output of the relay. For
includes test criteria, selection of test points, and test conditions. each zone impedance characteristic, three test points are
This work also explores the challenges of selecting test voltages considered by varying test angle. Three states namely, pre-
and currents considering impedance element directionality,
current magnitude supervision, and test equipment capacity
fault, fault/action, and reset are considered to mimic the fault
limitations. The calculations of voltage and current quantities in condition for each test point. Test criteria, selection of test
pre-fault, fault, and reset states are shown for each test point. points, calculation of AC test quantities along with the
The proposed methodology is implemented using Linear Ramp associated test conditions, and various constraints as well as
and Pulsed Linear Ramp algorithm. Finally, the test results are challenges are explained in the following Subsections.
discussed, and recommendations are made.
A. Test Criteria
Index Terms – Distance protection, Mho characteristic, An ideal relay should operate for a point inside of the
Quadrilateral characteristic, relay commissioning. operating boundary and should block the operation for a point
outside of that boundary. However, a practical relay has the
I. INTRODUCTION
tolerance margin due to hardware’s performance precision [2].
A power systems protection relay only operates few times The performance of the relay should be evaluated within the
in its whole service life. However, failure to operate as range of tolerance defined by relay manufacturer. The test
anticipated can result in wide-ranging damage, prolonged compares the theoretical expected value with the experimental
power failure or loss of life. Therefore, commissioning of value obtained to verify that the value is within the tolerance
relays before integration with the system and routine test levels.
during service life are regulatory requirements. Mechanical
inspection and functional test are two mandatory steps of relay
testing.
This paper emphasizes on the protection function test of
Mho and Quadrilateral distance element characteristics of the
modern digital distance relay. The distance element operating
characteristics are defined by reach, line angle, and
supervision conditions. Manufactures recommend testing a
relay at several given points inside and outside of the
characteristic. Nevertheless, the implementation of those
recommendations is not straightforward because of Figure 1: Three test points for Mho characteristic
supervision conditions and multifunction trip equation [1].
This paper shows a methodology on how to compute the fault
current and voltage quantities to trace the operating boundary B. Selection of Test Points
for a given angle subject to functional and testing hardware The test engineer/technician can select multiple test
constrains. The detail of the methodology is reviewed in points. More points will increase the calculation burden and
Section II. The implementation of the proposed methodology the testing time. That is why it is important to optimize the test
using Linear Ramp and Pulsed Linear Ramp algorithm is point selection. At least three points should be tested to assess
presented in Section III. Section IV covers the test results the performance of a non-linear characteristic. For Mho
along with recommendations for testing. Finally, concluding characteristic, the first test point is at maximum torque angle
remarks are made in Section V. and other two points are selected at two sides of maximum
torque angle as illustrated in Figure 1. The angles of the test
II. PROPOSED METHODOLOGY points are θL, θL+Δ and θL-Δ, respectively where θL represents
maximum torque angle. The value of Δ is user defined and can
The proposed methodology includes the procedure to test be varied to shift the test points. The impedance function is
the phase and ground distance function operating more exposed to fail at the sharp edges of the operating curve
due to overlapped tolerance area. That is why; the first test 5) Fault current is injected only in one phase for ground
point is selected at the edge to ensure the correct performance element tests.
of Quadrilateral characteristic as shown in Figure 2. Other two 6) Voltage drops only on faulted phases.
points are nominated by one from each side of the edge. The
edge angle θe is calculated based on the Equations (1-5). Ze, The calculations of the test current and voltage quantities
Xe, and Re are impedance, inductance, and resistance, in different states for each test point are illustrated below.
respectively for the edge. The tilt angle resulted from system Pre-fault State: In pre-fault state, phase currents are equal
non-homogeneity is represented by φ [3]. in magnitude and 120 degrees apart from each other. The
voltages are equal to rated voltage in magnitude and 120
degrees apart from each other. The default value and range are
mentioned in Table I.
TABLE I
DEFAULT VALUE AND RANGE OF CURRENT AND VOLTAGE IN PRE-FAULT STATE
Quantities Default Value Range
Phase current Nominal load 0 to nominal
current load current
Phase voltage Rated voltage -
Reset State: In reset state, phase currents are also equal in
magnitude and 120 degrees apart from each other. The
voltages in reset state are considered equal to zero. The default
Figure 2: Three test points for Quadrilateral characteristic value and range are mentioned in Table II.
TABLE II
= tan( ) (1) DEFAULT VALUE AND RANGE OF CURRENT AND VOLTAGE IN RESET STATE
Quantities Default Value Range
= tan( ) (2) Phase current 0 0 to nominal
load current
= (3) Phase voltage 0 -
Fault/Action State: In this state, the desired fault event is
= ∗ + (4) simulated on the zone operating boundary to check the
response of the relay. The fault event can be simulated either
= arctan( ) (5) by varying the fault voltages slowly while the fault currents
are kept constant; or by varying the fault currents keeping the
For three-phase fault, test points must be selected out of the fault voltages constant, until the operating boundary at a given
load encroachment area to avoid the blocking or reach needs point. In this paper, the fault currents are varied and the fault
to adjusted according to the load encroachment setting. voltages are kept constant. For each test point, the expected
fault voltages and currents depend on impedance reach and the
C. Calculations and Test Conditions type of fault. Equation (6) shows the relationship between
The calculations of current quantities depend on the test fault current (Ip), fault voltage (Vp), and impedance (Z) for
conditions. Modern distance protection relays are complex single phase fault.
multifunctional devices based on classical or advanced
operating principles that impose special requirements for their = = ∠ (6)
( ) ( )
testing. For multi-function relay output, it is important to
enable only the element which is selected to test. Fortunately,
modern relay facilitates element enable-disable function. In Equation (6), k is the zero-sequence compensation factor
Moreover, the test currents and voltages should be injected in and θ represents angle of the test point. For phase-phase fault,
such a manner so that the elements other than selected one the relationship between fault current (Ip), fault voltage (Vpp),
remain deactivated. In this paper, the test current and voltage and impedance (Z) can be expressed as Equation (7).
quantities are calculated based on the following test
conditions: = = ∠ (7)
1) Built-in element enable-disable option in the relay.
2) Three-phase voltage is applied for all three states.
3) Pre-fault and post-fault currents are injected in three Equation (8) shows the relationship between fault current (Ip),
phases. fault voltage (Vp), and impedance (Z) for three-phase fault.
4) Fault current is injected only in the selected phases
for phase element tests. = = ∠ (8)
reach value as shown in Figure 2. The calculations of Test
The expected fault current and fault voltage can be calculated Parameters for LR and PLR algorithms are shown in Table III
from Equations (6-8) for any desired fault event. In each and Table IV, respectively.
equation there are two unknown which include the fault
voltage and the fault current magnitude. In this paper, the
expected fault current (test current) magnitude is assumed
within given constraints to calculate the fault voltage for the
sake of mathematical simplicity.
D. Constraints and Challenges
The key challenge of the fault current and voltage
selection is to satisfy various constraints [2]. The arbitrary
assumption of fault current may result in the failure of the test
because of one or more unsatisfied constraints. The following
constraints must be considered to select the fault current for
any test point.
1) When the impedance element is supervised by the Figure 3: Working principle of LR algorithm for zone 1 test
fault current magnitude then the impedance element
is only picked up if the fault current is greater than a
specific threshold value.
2) The maximum possible value of the fault current
depends on the capacity of the Power System
Simulator used to simulate the fault event.
3) The practical fault voltage must be smaller than the
rated phase voltage. According to Equations (6-8),
the value of the fault voltage depends on the fault
current magnitude for a given test point. Therefore,
the fault current should be selected within a range so
that the fault voltage satisfies the constraint.
4) The ground impedance element is supervised by
directional conditions described in [2, 4]. The values
of the fault current and the fault voltage should be Figure 4: Working principle of PLR algorithm for zone 1 test
chosen to satisfy the conditions of ground
directionality.
TABLE III
TEST PARAMETERS CALCULATION FOR LR
III. IMPLEMENTATION Offset Current 50-95% of the Minimum Expected
The proposed test methodology can be implemented in Current of selected test points
various algorithms like State Simulator, Linear Ramp (LR) Offset Duration ≥5 cycle
and Pulsed Linear Ramp (PLR), etc. This paper only considers Delta Current >0.0 A (higher value requires higher delta
LR and PLR search algorithm. LR search algorithm changes time)
the current in single step to move from Pre-fault to Offset and Delta Time ≤(Time delay + 5) cycle
then it changes current in small steps until the element pickup. Current Limit 120% of Maximum Expected Current of
Each step size is denoted by Delta Current and Delta Time the selected test points
represents time for each step. The overview of LR working
principle for zone1 test is shown in Figure 3. TABLE IV
TEST PARAMETERS CALCULATION FOR PLR
PLR search algorithm also changes the current in single
Offset Current Load current
step to move from Pre-fault to Offset and then it uses current
Offset Duration ≥5 cycle
pulse instead of step to reach expected operating boundary. At
the end of each pulse, it moves to Offset point and stays for a Initial Pulse 90-95% of the Minimum Expected
time period which is called Wait Time. The overview of PLR Current Current of selected test points
working principle for zone1 test is shown in Figure 4 where Pulse Duration (Time delay + 5) cycle
the solid arrow lines indicate the trajectory of each pulse. The Wait Time ≥5 cycle
fault voltages for all test points are assumed to be equal and Delta Current >0.0 A (higher value requires higher delta
calculated based on the point which has highest impedance time)
reach value. For Mho characteristic, highest reach value is Current Limit 120% of Maximum Expected Current of
found at maximum torque angle. For Quadrilateral the selected test points
characteristic, the edge of the characteristic results in highest
IV. RESULTS AND DISCUSSIONS elements fail. The higher value of the Delta Current and lower
value of the Delta Time may address the issue in some extent.
The proposed test methodology is implemented in Linear
However, the Test Simulator requires higher delta time for
Ramp (LR) and Pulsed Linear Ramp (PLR) search algorithm
higher delta current.
and tested for single-function and multi-function outputs by
varying zone impedance reaches. The sample results are
shown in Figures 5-8. The results show that PLR successfully
operate within the tolerance range for both single-function and
multi-function relay output irrespective of zone reach.
However, LR only operates successfully within the tolerance
range for single-function relay output while frequently
operates out of the tolerance range for multi-function output.
Multi-function output consists of several zone elements as the
example shown in Equation (9).
= 1 2 3 (9)
Figure 6: Single-function test results for Quadrilateral
characteristic using LR algorithm (left) and PLR algorithm
(right)
On the other hand, PLR algorithm does not face above-
mentioned incorrect operations because PLR uses single big
jump as shown in Figure 4 to move from Offset to the
expected operating boundary in each pulse. The above feature
allows setting the offset point out of any zone operating area
and therefore, the timers of the zone elements get reset.
Figure 5: Single-function test results for Mho characteristics
using LR algorithm (left) and PLR algorithm (right)
The time delays of Zone1, Zone2, and Zone3 are T1, T2, and
T3, respectively and they are related as Equation (10).
< < (10)
As shown in Figure 3, for Zone1 test the impedance moves
from Pre-fault to Offset state which is within in the range of
Zone2 and finally it moves to the operating boundary of
Zone1 by multiple small jumps. If the required time to move
from Offset to operating boundary of Zone 1 is higher than T2,
Zone2 element will operate even before the impedance
reaches in Zone 1 operating boundary and subsequently, the
test will fail. This issue is more acute for Quadrilateral
characteristic. The higher value of Offset current (98% of the
Minimum Expected Current of selected test points) improves
the situation for single point test cases; although, it may result
Figure 7: Multi-function test results for Mho characteristics
in the pickup of the element in Offset state if the zone time
using LR algorithm (left) and PLR algorithm (right)
delay is small or zero. For same reason, all other zone
IV. CONCLUDING REMARKS
The paper proposes a methodology to test impedance
operating characteristics at multi-points of its boundary. The
article also explores two search algorithms to implement the
methodology in Power System Test Simulator. The analysis
suggests that Pulsed Linear Ramp (PLR) algorithm is much
appropriate one to implement the proposed impedance test
methodology while Linear Ramp (LR) can be limitedly
applied for single-element relay output test. The performance
of the proposed methodology implemented in LR algorithm is
vastly depends on the tuning of the test parameters such as
Offset current, Delta current, and Delta time.
REFERENCES
[1] M. Kezunovic, Y. Q. Xia, Y. Gao and et al, “An advanced
methodology for testing of distance relay operating
characteristic”, IEEE Transactions on Power Delivery, Vol.
11, No. 1, January 1996.
[2] Instruction Manual, SEL-421 Relay Protection and
Automation System, Schweitzer Engineering Laboratories Inc.
Figure 8: Single-function test results for Quadrilateral System, WA, USA, 2013.
characteristic using LR algorithm (left) and PLR algorithm [3] F. Calero, A. Guzman and G. Benmouyal, “Adaptive phase
(right) and ground quadrilateral distance elements”, 36th Annual
Western Protective Relay Conference, 2009.
[4] A. Polikoff, “Testing ground distance elements in SEL
distance relays”, SEL Application Guide, Vol. III, 2006.
