presents www.dlpe.gatech.

org

7th Annual
Fault and Disturbance Analysis
Conference

Distance Learning and Professional Education Sponsored by:

Georgia Institute of Technology School of Electrical and Computer Engineering
A Unit of the University System of Georgia Transient Recorder Users Council
www .truc.org
 Program Outline
Monday, April 26, 2004 Tuesday, April 27
8:00 Registration Morning Session
Global Learning & Conference Center Chairman: John Demko
8:30 Conference Opening Company: Arizona Public Service. Co.
Welcoming Remarks 8:00 Full Scale Experiment of the East-West Interconnection
A.P. Sakis Meliopoulos A. Grobovoy
Professor, School of Electrical and Computer Engineering Power System Emergency Control Laboratory Ltd.
Robert Orndorff A. Rabotin
Chairman, Transient Record Users Council Siberian ISO-/Interconnected Dispatcher Centre
Morning Session N. Lizalek
Chairman: Phillip L. Corliss Siberian Electric Power Research Institute
Company: Qualitrol Corp/Hathaway Inst. Div. M. Stubbe
8:40 Oscillation Detection and Recording Project Tractebel Energy Engineering
Dave Bertagnolli A. J. Germond
Xiaochuan Luo Ecole Polytechnique Federale de Lausanne
ISO - New England 8:40 Facilitating Restoration After System-Wide Disturbances
Jim Ingleson Using Automated Analysis
Joe Chow Mladen Kezunovic
Greg Allcorn Texas A&M University
New York ISO T. Popovic
Harish Mehta TLI, Inc.
Dick Dickens A. Chitambar
Jim Hackett Entergy Corp.
Mehta Tech B. Lunsford
Mark Kuras CenterPoint Energy
PJM Interconnection A. Bartylak
9:20 Enhancing Wide Area Information Systems in the Eastern ESKOM S.A
Interconnection 9:20 Fault Analysis as a Basic Input for Maintenance Management to
Matt Donnelly Improve the Overall Reliability of the Belgian Transmission System
Pacific Northwest National Laboratory P. De Cuyper
Paul Barber Elia Asets NV
Consortium for Electric Reliability Technology Solutions 10:00 BREAK
Quenton Hendricks Special Session: August 14, 2003 Blackout
Electric Power Group Moderator: Sakis Meliopoulos
10:00 BREAK Company: Georgia Institute of Technology
10:20 Distributed Web-based Systems for Recording of Wide Area 10:15 Disturbance Recording: Lessons Learned During Blackout of 2003
Disturbances Investigation
Alexander Apostolov Jim Ingleson
AREVA T&D Dean Ellis
11:00 Use of Phasor Measurements, SCADA and IED Data to New York ISO
Improve State Estimation Procedures 10:55 Role of Data Visualization in the Blackout 2003 Investigation
Mike Ingram, Sandra Bell, Sherica Matthews Tom Overbye
Tennessee Valley Authority University of Illinois, Urbana
A. P. Sakis Meliopoulos, George J. Cokkinides 11:30 Open Discussion
Georgia Institute of Technology 12:00 LUNCH
11:45 LUNCH
Afternoon Session
Afternoon Session Chairman: Tony Napikoski
Chairman: Harish Mehta Company: United Illuminating
Company: Mehta Tech, Inc. 1:00 Analysis of Under Voltage Load Shedding Event at Philadelphia,
1:10 Event Record Sampling Rates, Duration, and Fault Analysis; Actual Mississippi
Examples Meyer Kao and Gary Kobet
Jeffrey M. Pond Tennessee Valley Authority
National Grid USA Service Co., Inc. 1:40 Benefits of Power Swing Recording
1:40 Experiences Using Fault Recorders in the Peruvian Power System Andrew Klimek
Yofre Jacome NxtPhase Corporation
Red de Energia del Peru Robert Baldwin
Francisco Torres Southern California Edison
Comite de operacion economica del sistema 2:20 Managing Smart Device Communications with a Database
2:00 BREAK Robert M. Orndorff
2:20 Common Data Format Requirements for the Analysis of System Dominion Virginia Power
Disturbances 3:00 Digital Fault Recorder in Information Management Systems
Alexander Apostolov Xurui Wang, Xiaolong Sun, Yun Wang
AREVA T&D LEM China / Beijing Dianyan Electronic Instruments Co., Ltd.
3:00 Display and Discussion of Actual Fault Records Brought by 3:40 Improving Automatic Fault Analysis and Response Determination by
Participants the Use of Data Correlation from Multiple Sources, Including Non-
Robert C. Baldwin, Moderator Traditional, Non-Operational Data Sources
Southern California Edison David Kreiss
3:40 Users Forum William T. Shaw
Alan D. Baker, Moderator Kreiss-Johnson Technologies
Florida Power & Light Co. 4:20 Databases and DFR Records
4:30 ADJOURN Amir Makki
Softstuff, Inc.
5:00 ADJOURN
 EXPERIENCES USING FAULT RECORDERS IN THE PERUVIAN POWER SYSTEM

Yofr Jcome Depaz Francisco Torres Garca

Red de Energa del Per Comit de Operacin Econmica del Sistema

Abstract: This paper describes the experiences got in

the Peruvian power system using fault recorders, 2. THE PERUVIAN POWER SYSTEM
showing some records of dynamic and transient
phenomena and a brief analysis of them. The Peruvian power system has a radial configuration
fundamentally, with a transmission system in 220kV and
138kV, with a 2950 MW peak demand in the year 2003.
1. INTRODUCTION
The main load is concentrated in the center of the
Before in Peru, the fault analysis has been made using system (Lima) with predominantly hydraulic power
the protection relays (electro mechanicals, electronics, plants of generation very separated from the load, and
etc) indications and the event sequence recorded in the interconnected by very long lines.
control centers, with that information engineers could
identify the type of fault and the consequences. With
the Peruvian system growth came up dynamic
phenomena that originate the protection relays trip
without reason apparently.

The needed to explain the relays operations, obliged

some utilities to install basic fault recorders in
substations and lines, to get grater information, allow
identify: the type of fault, its location, magnitude, and
duration.

Nowadays with the technology development and

modernization of the Peruvian power system, the
control centers of some utilities have SCADA systems,
and remote access to digital fault recorders which are
synchronized with GPS system, giving more
information with time synchronization. With this
information is possible rebuilt the disturbances, analyze
the electric system behavior in order to determine the
origin of such and raising the solutions that are
necessary. Also, this information allow to verify the
performance of the protection relays and another
equipment of the power system.

The present Peruvian law imposes compensations in

case of loads interruption to the originator utilities of
the disturbance. With the purpose of determining the
responsibility of each utility, an analysis of each
interruption is made reconstructing the event and
determining the degree of participation of each utlity in
the event. The fault records and chronological sequence
Figure N 1. The Peruvian power system
of events, play a very important role in this analysis.

For this reason the utilities have increased the
implementation of digital fault recorders, in order to
demonstrate its degree of participation in the event,
being this fact very beneficial to know depth the
behavior of our electrical system.
These power plants present equivalent impedances in
terminal of the generators between 0,3 to 0,6 pu, for that
reason during dynamic conditions they have unstable
behaviors.
In the minimum load condition, the system presents
deficit of reactive inductive compensation, in the north
and south areas, become manifest with overvoltages.

1
Consequently, the Peruvian power system has critical
zones in which evidences dynamic and transitory
problems, being this The South Area, The North Area
and the Tingo Maria ring in the Center area.
3. FAULT RECORDS
In Peru at beginnings of years 80, were installed the
first electromechanical fault recorders in few
substations of the system, in order to register the
behavior of the main parameters of the power system
during a disturbance and later to be able to make the
respective analyses to determine the causes of the
disturbance. These equipments were denominated
Electromechanical Oscillograph, which printed the
record in special sheets with color fountain pen, as is
showed in the figure 2.
Figure N2. Electromechanical Oscillograph.
With those equipments and the signalings that gave the
electromechanical relays, It could be determined the
type of fault (single-phase, two phase or tree phase),
magnitude of the fault (Peak values) and from them the
analysis engineer calculated the phasor and plotted the
phasors diagrams.
They were very useful equipment however during its
operation they presented some problems, such as
- In severe events, they left the scale or the pen
remained joined
- The equipment was unfit.
- In summer the ink of the pen was spilled and in winter
it was congealed.
- High maintenance cost.
- Go to the place to gather the records.
The radial growth of the electrical system, the
interconnections with the North and South areas of the
country complicated the operation, appeared new
dynamic and transitory phenomena that no longer could
be registered with the electromechanical equipment,
being without explanation many of these events.
With the advance of the technology, nowadays exist in
the market digital fault recorders, that have additional
characteristics like: recording of data to variable
frequencies, capacity to record transitory events, storage
and reproduction of registered data, with the advantage
of the local and remote communication by means of a
personal computer. This characteristic allows to access
in few minutes and from remote places to the event data
and to extract the record, saving displacements of
specialized personnel to make this work.
In Peru, at the present time are installing in the main
lines of the system, digital fault recorders and digital
relays that have the function of oscillographic fault
record, with the purpose of identifying the origin of the
fault , to verify the performance of the protective
equipment, to locate the fault point and to take decisions
in real time.
The Figure N3, show the installation scheme which is
implementing. After happened a disturbance the
information stored in the memory of the digital fault
recorder, is transferred to a master station installed in the
control centers of the utilities, being this information
used by the operators to take decisions in real time.
Later, all the companies that have registered the event in
the diverse points of their facilities, analyze this
information and send it to the Committee of Economic
Operation of the System (COES), so that it makes the
respective analysis.
With the fault records got of the fault recorders or
protection relays, is come to make a fault analysis
observing the dynamic and transitory performance of the
electrical system, evaluating the frequency records,
digital records of the signals of the protection systems,
voltage profiles, magnitudes of currents and power.
From the analyses made, it has been possible to identify
dynamic phenomena like: Power swing, out-of-step,
voltage collapse; and transient electromagnetic
phenomena like: Resonance, ferroresonance, inrush and
troubles synchronizing power systems, some of these
cases are describe in this paper.
2
 Corriente en lneas L-213 y L-215 & Frecuencia

Printer Digital 800 60.6

Fault 700 60.4

Recorder 600 60.2

Local
V, I
Computer 500 60

Amperios

Hz
400 59.8

300 59.6

200 59.4

100 59.2

Interface 0 59

18:07:00
18:07:14
18:07:28
18:07:42
18:07:56
18:08:10
18:08:24
18:08:38
18:08:52
18:09:06
18:09:20
18:09:34
18:09:48
18:10:02
18:10:16
18:10:30
18:10:44
18:10:58
18:11:12
18:11:26
18:11:40
18:11:54
18:12:08
18:12:22
18:12:36
18:12:50
Hora

L-213 (A) L-215 (A) Frecuencia(Hz)

Modem Figure N5. Flow in the lines L-213 y L-215 & frequency of SEIN

(1) The increase of the electromechanical oscillations

Modem and the low damping that have the North area,
originated the disconnection of sensible loads to the
voltage variation, increasing the flow in the lines
aggravating the problem of power oscillations, that they
Printer took to this area to a lost of synchronism, which is
appraised in the record of the L-253 line.
Trigger
25/07/02
06:06:37 p..800

U1/kV

Master Control -100

-200
-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
t/s

Computer Center I1/A

Figure N3. System of digital fault recorder -500

-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
t/s

U2/kV

-100 -0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
t/s
-200

I2/A
3.1. Blackout produced by a small signal unstability 0

phenomena. -500
-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
t/s

U3/kV

The 25 of July 2002 at 18:10 hours, oscillations of Low I3/A

-100
-200
-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
t/s

frequency took place originated by the increase of 0 Figura N6. Oscilografa de la lnea L-253, S.E. Vizcarra.
-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

generation in the North area -500 t/s

MALACAS Figure N6. Record of line L-253, Vizcarra Substation

G
G AGUAYTIA

CARHUAQUERO

G TINGO MARIA
300

CHICLAYO
HUANUCO 275

2
250
L-121
225
GUADALUPE

3
VIZCARRA
1
SEIN 200

L-253
175
L-234 L-215 L-213
X/Ohm(Primary)

4 150

TRUJILLO 125
PARAMONGA

G 100

75
CANON DEL PATO
50

25
Figure N4. North area, Peruvian power system.
0

The increase of generation in the North area, is -25

appraised in the increase of the flow in the lines L-215 -150 -100 -50 0

R/Ohm(Primary)
50 100 150 200 250

(Chimbote Paramonga) and L-213 (Paramonga Z1 Z2 Z 3 Z5 Z L1E*

Huacho), being observed in the frequency that, as the

Figure N7. Impedance seen for the relay in the Vizcarra Substation
flow in these lines was increased began the
phenomenon of small signal power oscillation, as it is
appraised in figure 5. (2) The disconnection of the line L-253, caused the
disconnection of the line L-121 in the Tingo Maria
substation (figures N8 y 9) for out-of-step, due the

3
 1.0

Aguayta power plant stayed connected to the system 0.9

through a weak connection. 0.8

0.7

0.6

X/kOhm(Primary)
0.5

0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

-1000 -750 -500 -250 0 250 500 750 1000

R/Ohm(Primary)

Z1 Z2 Z3 LINEl Z L1E*

Figure N11. Impedance seen for the line L-234 relay, Guadalupe
substation.

120

110
Figure N8. Record of the line L-121, Tingo Mara Substation
100

90

80

70

X/Ohm(Primary)
60

50

40

30

20

10

-10

-20

-30

-75 -50 -25 0 25 50 75 100 125

R/Ohm(Primary)

Z1 Z2 Z3 LINEl Z L1E*

Figure N12. Zoom to the figure 11, Impedance in the operation

Figure N9. Voltage angle between Tingo Maria and Huanuco
Substations.
(3) In the North area, different combined modes
oscillation took place, identifying the local modes and
interarea modes, being this last one which produced an
out-of-step of this area in the Guadalupe substation,
which is why a line disconnected.
zones for the relay line L-234, Guadalupe substation.
(4) Finally the L-215 line disconnected by overvoltage,
producing a blackout in the north area.
3.2. A Voltage collapse phenomena
This event happened in the ElectroAndes area in august
14 of 2001, at 01:57 hours. In this event disconnected a
power autotransformer of 220/50kV in the Oroya
substation originating dynamic problems in that area
because the postfault configuration, producing a voltage
collapse.
G
AGUAYTIA

TINGO MARIA
ElectroAnde

PARAGSHA
HUANUCO

L-121 L-120 2

VIZCARRA
s

Figure N10. Record of the line L-234, Guadalupe Substation

78.8MW

1
During this event, disconnected the L-234 line SEIN OROYA
(Guadalupe Trujillo) in the Guadalupe substation
because the distance relay saw an out-of-step,
disconnecting the line. Figure N13. Area in which the voltage collapse was produce.

4
After the Oroya autotransformer disconnection, the
flow was increased in the Tingo Maria ring, and more
over due the connection is weak without reactive
compensation a great fall of tension was produced.
The L-120 line disconnection was produced by the
distance protection tripped, because the dynamic
impedance of the system entered to the operation zone
of the relay.

Figure N14. Record of L-120 line (Paragsha Huanuco), in

Paragsha substation.

When the records where analyzed, it was observed that
it did not correspond to a loss of synchronism, reason
why from the voltage and current obtained from the
record, the active power was calculed for this line,
according to the figure 15.
Figure N17: Impedance and relay zone of the L-120 line, paragsha
substation, plotted from the fault record.
3.3. A Resonance phenomena
The resonance event happened in December 23 of 2001
at 09:04 hours and occurred in the Tingo Maria ring,
which was in radial configuration, the 220kV lines were
with low load. That configuration was produced because
a line link of the ring was on maintenance, The L-253
(Vizcarra Paramonga) and the SVC (+90/-45 MVAr)
of the Vizcarra substation, as is shown in the figure
N18.

L-251 S .E.
HU A N U C O
C.T
AGUAYTIA

S .E.
TIN GO SISTEMA
M A R IA ELECTRO
S .E. ANDES
AUCAYACU
SVC
A N TA M IN A

Figure N15: Voltage, current and active power, L-120 line.

18 MW S.E
VIZCARRA

From the previous record, some points of voltage and

active power were taken, those points were plotted in L-253

the plane P-V, getting as a result a characteristic voltage SI ST E MA

I NT E RCONE CT AD O
collapse, the line disconnection occurred just at the Lnea en NACI O NAL

lowest voltage point mantenimiento

Figure N 18. Pre-fault configuration.

During the equipment maintenance, the bus bars of this

area were operating with a voltage in the range of
242kV because there were not enough reactive control
elements. The SVC maintenance in the Vizcarra
substation had finalized before the line, in order to
manage voltage and take load in Antamina, the SVC
entered in service, taking place oscillations of voltage
and current because of this operation, choosing the
operator to disconnect the SVC immediately.

Figure N16: Voltage, current and active power, L-120 line.

5
The System Coordinator after verified the stable
conditions of the system and didnt have clear
evidences of the oscillations source, ordered the SVC
connection, taking place again oscillations of voltage
and current (Figure N19), the operator of the Vizcarra
substation disconnected the SVC immediately, stopping
therefore the oscillations.
Figure N 19. Oscilografic record of a Ferroresonance at industrial
frequency in the Tingo Maria ring.
A third attempt was made, with the same previous
results, that is why the system coordinator of the
System arranged that the SVC stayed out of service
until the L-253 line enter in operation.
Later, the oscillographic records were analyzed and it
was observed that those records had a characteristic of
maximum current and maximu m voltage, but in other
time both variables were minimum. This characteristic
gave us the idea of a resistive load, but that only happen
in resonance, that resonance should give us a high
voltage without oscillations. However, the oscillatory
characteristic took to us to analyze in detail the
components of the network, reaching the following
conclusions:
* The topology that presents the Tingo Maria ring for
the L-253 disconnection (open ring) and operating
with low load, takes this area of the system to an
operation point near the condition of resonance at
industrial frequency (60Hz).
* The voltage in this area without compensation
elements, is not controllable and reach to have high
values (1.1 Pu), conditioning it to the appearance of
electromagnetic phenomena.
* The SVC connection in the Vizcarra substation under
that conditions contributes to the resonant component
of the network, and later when give more reactive
inductive component removes it from that condition.
Therefore, the oscillating resonant behaviors was
produced by the interaction of the SVC under the high
voltages that appeared.
It was possible to determine the cause of the strange
behavior of this area, aided by the oscillographics
records, which show us the behavior of the system
variables and the analysis elements to understand the
phenomena.
3.4. Synchronization problems.
The synchronization problems in Peru have appeared
when carrying out maneuvers in manual form and by
fault in the synchronism equipment, causing a strong
disturbance in the system and giving as result the
immediate disconnection or stay in service with
oscillations
One of this events happened on Monday, December 23
of 2001 at 10:54:55 hours in the Tingo Maria ring, when
a line of this ring was on maintenance. The event began
with the breaker opening in the autotransformer of
50MVA and 220/138kV in the Tingo Maria substation,
because the overvoltage protection tripped, in that
condition the Aguaytia area were separated of the
system (figure 20).
AREA SIST EM A
AISLADA S.E. TINGO
MARA
INTERCONECTADO
Lnea en
mantenimiento
Figure N 20 Diagram of the false synchronization.
In that condition, the operator thought that the Aguaytia
system was connected to the system trought the L-253
line which was on maintenance, closing the breaker
without synchronism check, originating a strong
disturbance with an oscillation of 1.5 seconds, that
ended with the Aguaytia Power plant disconnection, the
loads of this area finally, remained connected to the
system.
The strong disturbance, was caused by the closing of the
breaker with a great angular difference (60), appearing
suddenly a great current with a continuous component,
like a three phase short circuit, followed by a slow
oscillation when the aguaytia power plant tried to
resynchronized to the system, disappearing when the
aguaytia power plant tripped by the loss of synchronism.
The synchronizing equipment sent the signal to close the
breaker, when being with inadequate settings because in
that point synchronization maneuvers were not made.
Using the oscillographic record shown in the figure
N21, got in the L-121 line en the Tingo Maria
substation, it was possible determinate what happened in
this event.
6

Paralelo de la Prdida de paso de
CT. Aguayta la CT. Aguayta
1 1/2 segundos
Figure N 21. Oscillography of the synchronization between
Aguaytia area with the SEIN.
Another event happened on January 24 of 2004, at
03:39:10 hours, when were synchronized two areas the
South area with the Center-north area after a fault in the
interconnection line which separated the system.
After the fault in the L-2054 line, this line was
energized in the Cotaruse substation in order to
synchronize automatically in the Socabaya substation.
However, when variations of frequency appearing in
the South area, the synchronism relay did not allow the
closing of the breaker. That is why the synchronism
was made in manual mode.
Figure N 22. Oscillography of the effects of a forced
synchronism between the south and the Center-
North area in the Socabaya substation.
The operator saw that the conditions were very variable
and wait until he saw conditions to close the breaker,
causing a disturbance with power oscillations with a
frequency of 1.25 Hz in the power system, which were
attenuated in 5 seconds according to the figure N22.
The disturbance caused a voltage drop of 20% in the
busbars near the socabaya substation during 600ms
approximately, after that it recovered (figure N23), but
that voltage drop tripped some industrial loads by under
voltage protection.
Figure N 23. Voltage oscillography in the 220kV busbar, Ilo2
substation.
The disturbance was caused when the synchronization
was made with inadequate parameters. With the
oscillography shown in the figure 24, it could be
determinate the parameters in the synchronization
between both systems, been this values the following: f
= 0.2 Hz; V = 2.49 kV y = 91.
Figure N 24. Analysis of the synchronization of the South operative
area to the center-north using oscillographic records.
From these values, we can conclude that the angle
between power systems in the synchronization was too
big, for that reason during the closing, appeared a high
current with a voltage drop, for the power system it was
as a three-phase short circuit, but the system recover the
stable condition after 5 seconds, without generation
shedding.
As it was shown, using oscillographic records and
software available for the analysis it is possible to
determinate the origin of the problems and the effects in
the power system.
4. CONCLUTIONS
The oscillographic records nowadays are a tool of
extreme importance for the electrical power system
7
analysis; the application of these recorders in Peru has
allowed detecting dynamic and electromagnetic
transients phenomena, determining its origin and
raising the remedial actions that allow to control or to
eliminate the effects of these phenomena.
In this paper, we showed some cases in which
oscillographic records were obtained that allowed us to
make the analyses of the faults and events in the
electrical system.
Another application that has in Peru is to evaluate the
performance of the protection relays, sending to the
relays the digital signals that are analyzed in the
performance during faults. Also, the oscillographic
records obtained in the events, are used to verify and to
improve the protection relays settings. This verification
of settings is made in the protection laboratory, using
relay testers, the record is save in Comtrade format and
then is injected into the relay observing the
performance, and varying the settings according to the
requirement of the case.
6. AUTHORS
Francisco Torres Garca Ing , MSc , was born in
LimaPer, He studied in the National Engineering University,
receiving the Dipl. Eng. Degree in 1985. He received M.Sc
degree from the Universidad Federal de Ro de Janeiro -
Brasil, in 1997. Currently, he works for the Comit de
Operacin Econmica del Sistema Peruano (COES) as a
Specialist in Supervision of the Quality and he is profesor at
Nacional Engineering University.
Yofr Jcome Depaz, was born in HuarazPer, He
studied in the Nacional Engineering University, receiving the
B.S. degree in 1998. Currently he Works as Specialist in
Analysis and protection of power systems in the Company Red
de Energia del Peru (REP).
Gratefulness
We thank for to the Comite de Operacion Economica
del sistema (COES) and the utilities members of this by
the support and facilities offered for the development of
the present article.

Finally, we comment that at these moments in the

Peruvian power system are being implemented in
stages, digital fault recorders in the lines of 138 and 220
kV and important power plants of generation, with the
purpose of being able to observe the dynamic and
transitory behavior of all the components in the
electrical system.

5. REFERENS
[1] Kundur, P. "Power System Stability and control"
Power System Engineering Series. McGraw-Hill,
1994
[2] Anderson, P.M. and Fouad, A. A. "Power System
Control and Stability" IEEE Press Power
Engineering Series, 2003.
[3] Van Cutsem, T. And Vournas. "Voltage Stability
of Electric Power System". Power Electronics and
Power Systems Series. Kluwer Academic
Publishers, 1998.

8
 EXPERIENCES USING FAULT RECORDERS
IN THE PERUVIAN POWER SYSTEM

7th Annual Fault and Disturbance Analysis Conference

April 26-27, 2004 Atlanta, Georgia
 FAULT RECORDERS

Previously fault analysis was based on relay

indications and event sequences recorded in
control centers.
The expansion of the Peruvian power system
caused dynamic phenomena resulting in
protection relays trips without apparent reason.
It was necessary to install basic fault recorders
to gather more information to explain the relays
operations.
F.Torres & Y. Jacome 26-27 April, 2004 2
 FAULT RECORDERS

With the advancement of technology we now

have digital recorders with remote access,
synchronized with GPS.
With the information obtained from this digital
fault recorders is possible reconstruct the
disturbances, analyze the electric system
behavior, verify the performance of the
protection relays.

F.Torres & Y. Jacome 26-27 April, 2004 3

 FAULT RECORDERS

The Peruvian law mandates compensation to

consumers in case of load interruptions.
For each interruption, an analysis is made to
reconstruct the event and determine the
responsibilities.
Power companies in Peru have increased the
implementation of digital fault recorders, to help
determine / defend its degree of participation in
an event.
F.Torres & Y. Jacome 26-27 April, 2004 4
 THE PERUVIAN POWER SYSTEM

Radial configuration: Lima at the center,

A transmission system in 220 kV and 138 kV,
Load is concentrated in the center of the system,
Generation: Primarily Hydro (80-85%) located at great
distances from load concentrations,
Equivalent impedances at Power Plants are between 0.3 to
0.6 pu. Unstable behavior under dynamic conditions,
Sustained over-voltages under minimum load conditions.

F.Torres & Y. Jacome 26-27 April, 2004 5

 THE PERUVIAN POWER SYSTEM

Northern Region

Central Region

Southern Region

F.Torres & Y. Jacome 26-27 April, 2004 6

 THE PERUVIAN POWER SYSTEM

The Ring of
Tingo Mara

F.Torres & Y. Jacome 26-27 April, 2004 7

 FAULT RECORDERS

The electromechanical
fault recorders were
difficult to maintain
High maintenance cost.
For each event, people
had to travel to each
sub-station to gather
the records

F.Torres & Y. Jacome 26-27 April, 2004 8

 FAULT RECORDER SYSTEM
Printer Digital
Fault
Recorder
Local
Computer V, I

Interface

Configuration of a typical
Digital Fault Recorder System. Modem

Modem

Printer

Master Control
Computer Center

F.Torres & Y. Jacome 26-27 April, 2004 9

 FAULT RECORDS ANALYSIS

We will examine data and analysis of real cases

involving,
Blackout produced by a small signal instability
phenomenon,
A voltage collapse phenomenon,
A resonance phenomenon,
Synchronization problems.

F.Torres & Y. Jacome 26-27 April, 2004 10

 Blackout produced
Blackout produced byby
aa small
small signal
signal instability
instability phenomenon
phenomenon

F.Torres & Y. Jacome 26-27 April, 2004 11

 Blackout Produced by a small
Signal instability phenomenon

MALACAS

The increase in generation G

G
AGUAYTIA

in the northern region CARHUAQUERO

caused low frequency G TINGO MARIA

oscillations. CHICLAYO
HUANUCO

L-121

GUADALUPE VIZCARRA
SEIN
L-253
L-234 L-215 L-213

TRUJILLO
PARAMONGA

CANON DEL PATO

F.Torres & Y. Jacome 26-27 April, 2004 12

 Blackout Produced by a small
Signal instability phenomenon

Corriente en lneas L-213 y L-215 & Frecuencia

MALACAS

800 60.6 G
G AGUAYTIA
700 60.4
CARHUAQUERO
600 60.2
TINGO MARIA
G
500 60
CHICLAYO HUANUCO
Amperios

Hz

Hz
400 59.8
L-121
300 59.6
GUADALUPE
200 59.4
VIZCARRA
SEIN
L-253
100 59.2
L-234 L-215 L-213
0 59
18:07:00
18:07:14
18:07:28
18:07:42
18:07:56
18:08:10
18:08:24
18:08:38
18:08:52
18:09:06
18:09:20
18:09:34
18:09:48
18:10:02
18:10:16
18:10:30
18:10:44
18:10:58
18:11:12
18:11:26
18:11:40
18:11:54
18:12:08
18:12:22
18:12:36
18:12:50 TRUJILLO
PARAMONGA
Hora

L-213 (A) L-215 (A) Frecuencia(Hz)

G
CAON DEL PATO

The increase of the electromechanical oscillations and the low damping

in the northern region, caused voltage variations resulting in
interruption in operation of sensitive electrical equipment.
F.Torres & Y. Jacome 26-27 April, 2004 13
 Blackout Produced by a small
Signal instability phenomenon

MALACAS
G
G AGUAYTIA

CARHUAQUERO

TINGO MARIA
G
CHICLAYO HUANUCO

L-121
Hz
GUADALUPE VIZCARRA
SEIN
L-253
L-234 L-215 L-213
Lose synchronization
TRUJILLO
PARAMONGA

G
CANON DEL PATO

The drastic change in loads aggravated the problem of power

oscillations, causing this region to lose synchronization.

F.Torres & Y. Jacome 26-27 April, 2004 14

 Blackout Produced by a small
Signal instability phenomenon
Trigger
25/07/02
06:06:37 p..800

U1/kV

0
-100 -0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 t/s
-200
I1/A

0 MALACAS
-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 t/s
-500

U2/kV
G
0
G AGUAYTIA
-100 -0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
t/s
-200

I2/A CARHUAQUERO
0
-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
TINGO MARIA
-500

U3/kV
t/s
G
0 CHICLAYO
-100 -0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
t/s
HUANUCO
-200

I3/A

0
-0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
L-121
-500 t/s

GUADALUPE VIZCARRA 1
SEIN
L-253

300
L-234 L-215 L-213
275

250

TRUJILLO
225
PARAMONGA
200

175
G
X/Ohm(Primary)

150

125 CANON DEL PATO

100

The loss of synchronization caused a

75

50

25

-25

-150 -100 -50 0 50 100 150 200 250

trip on the line L-253.
R/Ohm(Primary)

Z1 Z2 Z3 Z5 Z L1E*

F.Torres & Y. Jacome 26-27 April, 2004 15

 Blackout Produced by a small
Signal instability phenomenon

MALACAS
G
G AGUAYTIA

CARHUAQUERO

TINGO MARIA
G
CHICLAYO HUANUCO

L-121
2
GUADALUPE VIZCARRA 1
SEIN
L-253
L-234 L-215 L-213

TRUJILLO
PARAMONGA

G
CANON DEL PATO

The line L-253 trip caused line L-121

to trip due to out-of-step condition.

F.Torres & Y. Jacome 26-27 April, 2004 16

 Blackout Produced by a small
Signal instability phenomenon

Loss of Stability at Aguayta Thermal Power Plant

Tingo Mara
L-251 L-121 L-120
73.2km 88.16km 86.21km SEIN
C.T.Aguayta Hunuco
E2 P E1 0

XL

E1 E2
P= Sen
XL
Static
1.2
Stability
1.0
limit

Potencia (p.u)
0.8

0.6

0.4

0.2

0.0

30

60

120

150

180

210
90
Angulo

F.Torres & Y. Jacome 26-27 April, 2004 17

 Blackout Produced by a small
Signal instability phenomenon
OSCILOGRAFA EN LNEA L104 - S.E. CHIMBOTE 1

f = 1.33 Hz (modo local de oscilacin)

MALACAS
f = 3.2 Hz (modo inestable de oscilacin) G
G AGUAYTIA

f = 2.1 Hz (modo local de oscilacin) CARHUAQUERO

TINGO MARIA
G
CHICLAYO HUANUCO

L-121
2
GUADALUPE VIZCARRA 1
SEIN
L-253
L-234 L-215 L-213

TRUJILLO
PARAMONGA

G
In the northern region, different CANON DEL PATO

oscillation modes were observed

F.Torres & Y. Jacome 26-27 April, 2004 18

 Blackout Produced by a small
Signal instability phenomenon

MALACAS
G
G AGUAYTIA

CARHUAQUERO

TINGO MARIA
G
CHICLAYO HUANUCO

3
L-121
2
GUADALUPE VIZCARRA 1
SEIN
L-253
1.0

L-234 L-215 L-213

0.9

0.8

0.7
TRUJILLO
0.6 PARAMONGA
X/kOhm(Primary)

0.5

0.4 G
0.3

CANON DEL PATO

0.2

0.1

One of this modes produced an Out-

0.0

-0.1

-0.2

-0.3

-1000 -750 -500 -250 0

R/Ohm(Primary)
250 500 750 1000
of-step, and tripped the line L-234
Z1 Z2 Z3 LINEl Z L1E*

F.Torres & Y. Jacome 26-27 April, 2004 19

 Blackout Produced by a small
Signal instability phenomenon

120

110

100

90

80

70
X/Ohm(Primary)

60

50

40

30

20

10

-10

-20

-30

-75 -50 -25 0 25 50 75 100 125

R/Ohm(Primary)

Z1 Z2 Z3 LINEl Z L1E*

F.Torres & Y. Jacome 26-27 April, 2004 20

 Blackout Produced by a small
Signal instability phenomenon

The voltage in the Substations MALACAS

G
G
AGUAYTIA

increased, tripping line L-215 CARHUAQUERO

TINGO MARIA
G
CHICLAYO HUANUCO

3
L-121
2
GUADALUPE VIZCARRA 1
SEIN
L-253
L-234 L-215 L-213

4
TRUJILLO
PARAMONGA

G
CANON DEL PATO

F.Torres & Y. Jacome 26-27 April, 2004 21

 Blackout Produced by a small
Signal instability phenomenon

Finally there was a blackout of MALACAS

G
G
AGUAYTIA

The northern region CARHUAQUERO

TINGO MARIA
G
CHICLAYO HUANUCO

3
L-121
2
GUADALUPE VIZCARRA 1
SEIN
L-253
L-234 L-215 L-213

4
TRUJILLO
PARAMONGA

G
CANON DEL PATO

F.Torres & Y. Jacome 26-27 April, 2004 22

 A voltage
A voltage collapse
collapse phenomenon
phenomenon

F.Torres & Y. Jacome 26-27 April, 2004 23

 A Voltage collapse phenomenon

G
AGUAYTIA
This event happened in the
ELECTROANDES area
TINGO MARIA

ElectroAnde
PARAGSHA
HUANUCO

L-121 L-120
20 MW

s
VIZCARRA

78.8MW
1

SEIN OROYA

F.Torres & Y. Jacome 26-27 April, 2004 24

 A Voltage collapse phenomenon

G
AGUAYTIA
A power transformer
disconnected in the Oroya
TINGO MARIA
substation, producing a

ElectroAnde
PARAGSHA
voltage collapse. HUANUCO

L-121 L-120

60 MW

s
The flow was increased in VIZCARRA

the Tingo Mara ring,

1
producing a great fall of
voltage.
SEIN OROYA

F.Torres & Y. Jacome 26-27 April, 2004 25

 A Voltage collapse phenomenon

G
AGUAYTIA

TINGO MARIA

ElectroAnde
PARAGSHA
HUANUCO
2
L-121 L-120

60 MW

s
VIZCARRA

SEIN OROYA

F.Torres & Y. Jacome 26-27 April, 2004 26

 A Voltage collapse phenomenon
Active and Reactive Power in the line L-120

Punto5 : Punto6 :
Punto4 : Punto1 : Punto2
Vf = 40.638 VfkV= 28.977
:Punto3kV
Vf =kV
Vf = 50.138 64.999 kV
MW = 22.639 :
Vf
MW = 31.461 = 61.178 kV
Vf = 56.989 kV
MW = 38.537
MW = 46.084 MW = 52.077MW = 54.894

F.Torres & Y. Jacome 26-27 April, 2004 27

 A Voltage collapse phenomenon

P1
P2

P3

P4

P5

P6

F.Torres & Y. Jacome 26-27 April, 2004 28

 A Voltage collapse phenomenon

G
AGUAYTIA

TINGO MARIA

ElectroAnde
PARAGSHA
HUANUCO
2
L-121 L-120

60 MW

s
VIZCARRA

SEIN OROYA

F.Torres & Y. Jacome 26-27 April, 2004 29

 A resonance
A resonance phenomenon
phenomenon

F.Torres & Y. Jacome 26-27 April, 2004 30

 A Resonance phenomenon
L-251
IN-2344
IN-2342 L-121 L-120
40MVA IN-4078

C.T IN-4068 IN-4064 IN-4062 IN-4060 S.E.. PARAGSHA

S.E PARAGSHA 2
AGUAYTIA
IN-2340
S.E. TINGO
S.E
30 MVAR
MARIA 138 KV
IN-2346 MARIA S.E.
S.E. HUANUCO
HUANUCO
SISTEMA
ELECTROANDES
S.E .
S.E.
AUCAYACU
AUCAYACU
L- 252
SVC

IN-2404 IN-2406
ANTAMINA
ANTAMINA
IN-2408 S.E.. TOCACHE
S.E L-224
L- 255
BARRA 220kV
242 kV
S.E
VIZCARRA
VIZCARRA
220KV IN-2402 IN-2400
18MW 220 kV
SISTEMA
SISTEMA
CENTRO
CENTRO
SUR
SUR
220 kV
40
MVAR
L- 253
IN-2224 S.E.
S.E.
ZAPALLAL
ZAPALLAL
IN-2338
S.E .
IN-2320
SISTEMA
SISTEMA 220KV PARAMONGA
PARAMONGA
NUEVA
NORTE
NORTE IN-2198
IN-2192 IN-2390 IN-2388
L-215 IN-2194

IN-2196 L-213 L-212

IN-2200

CAHU
CAHU HUACHO
A
A

This event happened in the ring of Tingo Mara

The SVC and the L-253 were on maintenance
The voltage in the Vizcarra substations was 1.1pu
F.Torres & Y. Jacome 26-27 April, 2004 31
 A Resonance phenomenon
L-251
IN-2344
IN-2342 L-120
L-121
IN-4078
40MVA
C.T IN-4068 IN-4064 I N-4062 IN-4060 S.E.. PARAGSHA 2
S.E
AGUAYTIA
IN-2340
S.E.. TINGO
S.E TINGO
30 MVAR
MARIA 138 KV
IN-2346
MARIA S.E .. HUANUCO
S.E HUANUCO
SISTEMA
ELECTROANDES
Beginning of S.E.
S.E.
the event L-252
AUCAYACU
AUCAYACU

SVC

IN-2404 I N-2406
ANTAMINA
ANTAMINA
IN-2408 L-224
L-255 S.E.. TOCACHE
BARRA 220kV 242 kV
S.E
VIZCARRA
VIZCARRA
220KV I N-2402 IN-2400
18MW 220 kV

RESONANCE SISTEMA
CENTRO
CENTRO
SUR
220 kV
40
MVAR
L-253
IN-2224 S.E.
S.E.
ZAPALLAL
IN-2338 S.E.
SISTEMA 220KV IN-2320
PARAMONGA
NUEVA
NUEVA
NORTE
NORTE IN-2198
IN-2192 IN-2390 I N-2388
L-215 IN-2194

IN-2196 L-213 L-212

I N-2200

CAHU HUACHO
A

In order to manage the voltage in the Vizcarra substation, the SVC was
entered into service, resulting in oscillations of voltage and current.
F.Torres & Y. Jacome 26-27 April, 2004 32
 A Resonance phenomenon

The records from Tingo Maria substations show maximum voltage

and maximum current characteristics, but at other times both were
minimum.
F.Torres & Y. Jacome 26-27 April, 2004 33
 A Resonance phenomenon
L 255 L 252
55 km 152 km S.E. TingoMara

svc
S.E. Vizcarra

9 MVAr 24 MVAr 30 MVAr

Capacitive Capacitive Inductive

Equivalent resonant circuit.

F.Torres & Y. Jacome 26-27 April, 2004 34

 A Resonance phenomenon

953

59 Hz

Harmonic impedance under the conditions the ring of Tingo Mara open.

F.Torres & Y. Jacome 26-27 April, 2004 35

 Synchronization problems
Synchronization problems

F.Torres & Y. Jacome 26-27 April, 2004 36

 Synchronization problems: Case 1

AREA
SISTEMA
AISLADA
INTERCONECTADO
S.E. TINGO MARA

Line on
maintenance

This event happened in the Tingo Mara Ring,

A line was out of service in maintenance,
A breaker was tripped by the overvoltage protection, isolating the
Aguayta area.

F.Torres & Y. Jacome 26-27 April, 2004 37

 Synchronization problems: Case 1
Prdida de paso de
Paralelo de la
CT. Aguayta la CT. Aguayta
1 1/2 segundos

The operator closed the breaker without synchronism check

A strong disturbance was resulted, that caused the disconnection
of Aguaytia power plant.

F.Torres & Y. Jacome 26-27 April, 2004 38

 Synchronization problems: Case 2

A fault in the interconnection line

between the Central/Northern
regions and the Southern region,
caused the isolation of the Southern
region

F.Torres & Y. Jacome 26-27 April, 2004 39

 Synchronization problems: Case 2

After the fault the line was

energized in order to synchronize
automatically in the Socabaya
substation
The synchronization relay did not
allow the closing of the breaker in
the Socabaya substation
The substation operator forced
synchronization in manual mode

F.Torres & Y. Jacome 26-27 April, 2004 40

 Synchronization problems: Case 2

The disturbance caused

power oscillations with
a frequency of 1.25 Hz
in the power system,
which were attenuated
in 5 sec.

F.Torres & Y. Jacome 26-27 April, 2004 41

 Synchronization problems: Case 2

The disturbance caused a

voltage drop of 20% in the
busbars near the Socabaya
Substation, during 600ms
approximately.

F.Torres & Y. Jacome 26-27 April, 2004 42

 Synchronization problems: Case 2

The disturbance happened

because the synchronization
was forced with an
inadequate value for the
Angle parameter.
Using the fault records it
was possible determinate the
synchronization values.
f = 0.2 Hz
V= 2.49 kV
= 91

F.Torres & Y. Jacome 26-27 April, 2004 43

 CONCLUSIONS

Digital fault recorders nowadays are a tool of extreme

importance for electrical power system analysis.
The use of digital recorders in Peru, has allowed the
detection of dynamic and electromagnetic transient
phenomena.
Digital fault recorders and the Comtrade Format is
used to evaluate the performance of the protection
relays.

F.Torres & Y. Jacome 26-27 April, 2004 44

 THANK YOU
FOR YOUR ATTENTION

F.Torres & Y. Jacome 26-27 April, 2004 45