Fault Location Principles

7/28/2019 Fault Location Principles

1/13

ABBGroupMarch26,2012|Slide1

Fault Location

Principles

Dr. MURARI MOHAN SAHAABB AB

Vsters, SwedenKTH/EH2740 Lecture 4


Dr. Murari Mohan Saha was born i n 1947 in Bangladesh. He receivedB.Sc.E.E. from Bangladesh Universit y of Technology (BUET), Dhaka in 1968

and compl eted M.Sc.E.E. in 1970. During 1969-1971, he was a lect urer at theE.E. dept.,BUET. In 1972 he compl eted M.S.E.E and in 1975 he was awardedwith Ph.D. from The Technical Universit y of Warsaw, Poland. He joined ASEA,Sweden in 1975 as a Development Engineer and currently is a SeniorResearch and Development Engineer at AB B AB, Vsters, Sweden. He is a

Senior Member of IEEE (USA) and a Fellow of IET (UK). He is a r egisteredEuropean Engineer (EUR ING) and a Chartered Engineer (CEng). His areas of

interest are measuring transformers, power system analysis and simulation,and digital protective relays. He holds 35 granted patents and produces morethan 200 technical papers. He is the co-author of a book, entit led, Faultlocation on Power Networks , published by Springer, January 2010.

Presenter


Contents

Introduction

One-end fault location

Two-end/Multiterminal fault location

Fault location on distribution networks

Conclusions

Information aboutbookon FaultLocation


Introduction


It is a device or apparatus placed at one end of a station, which displays thedistance to fault (in km or in % of line) following a fault in a transmission line.

ZA ZB

ZL

Line

Relay

FaultLocator

Line

Relay

Line section length

Fault distance

Introduction What is a Fault Locator?


Introduction

When a fault occurs on a line (distribution ortransmission), it is very important for the utility toidentify the fault location as quickly as possible forimproving the service reliability.

If a fault location cannot be identified quickly and thisproduces prolonged line outage during a period ofpeak load, severe economic losses may occur andreliability of service may be questioned.

All these circumstances have raised the greatimportance of fault-location research studies and thusthe problem has attracted widespread attention amongresearchers in power-system technology in recentyears.


2/13


Introduction

Fault location is a process aimed at locating the

occurred fault with the highest possibly accuracy.

Fau lt l oc ator is mainly the supplementaryprotection equipment, which apply the fault-locationalgorithms for estimating the distance to fault.

When locating faults on the line consisting of morethan one section, i.e., in the case of a three-terminalor multi-terminal line, the faulted section has to beidentified and a fault on this section has to be located.


Introduction

A fault-location function can be implemented into:

microprocessor-based protective relays

digital fault recorders (DFRs)

stand-alone fault locators

post-fault analysis programs


Introduction

Fault locators versus protective relays differences related to the following features:

accuracy of fault location

speed of determining the fault position

speed of transmitting data from remote site

used data window

digital filtering of input signals and complexityof calculations


Introduction

General division of fault location techniques:

technique based on fundamental-frequency currentsand voltages mainly on impedance measurement

technique based on traveling-wave phenomenon

technique based on high-frequency componentsof currents and voltages generated by faults

knowledge-based approaches

unconventional techniques (fault indicators installed either insubstations or on towers along the line; monitoring transients of

induced radiation from power-system arcing faults using both VLF

and VHF reception )


Voltage & Current MeasurementChains


Voltage & Current Measurement Chains

CURRENT

TRANSFORMERS

vp

ip

vs

is

v2(n)

i2(n)

POWER

SYSTEM

CTs

VTsMatching

Transformers

MatchingTransformers

Analogu eFilters

Analogu e

Filters

A/D

A/D


3/13


4/13

First Stand Alone NumericalFault Locator on Commercial Use

where:

FFLAA RIpZIU

A

FAF

D

II

SBLSA

SBLA

ZZZ

Zp)Z-(1D

EA

pZL

Fault Locator

Line section length

Fault distance

EB

ZSA ZSB(1-p)ZL

RF

IB

IA

IF

A B

One-end Fault Location Algorithm Compensatingfor Remote End Infeed Effect

where:

FA

FALAA R

D

IpZIU

L

SB

LA

A1

Z

Z

ZI

UK 1

0RKKpKp F3212

L

SB

LA

A2

Z

Z

ZI

UK 1

L

SBSA

LA

FA3

Z

ZZ

ZI

IK 1

One-end Fault Location Algorithm Compensatingfor Remote End Infeed Effect

where:

OAPOMFA

FALAA IZR

D

IpZIU

LSBSA

SBLSBSAA

ZZ2Z

ZZZp)(Z-(1D

2

)

ZL

ZSA

ZSB

pZL

FL

FL

P

IOAP

ZOM

RF

(1-p)ZL

One-end Fault LocationAlgorithm Compensating forRemote End InfeedEffect Case of Parallel Lines

Relay input Input transformers

Filter low pass

Multiplexer

Hold circuit

Analog/digital converter

Micro processor

Telemeter outputLed-indykator

Parameter setting

Data and program memory

Peripheral interface adapter

Printer output

Input signals from:Line protectionTrip Phase selection Currents Voltages

Collection of I0

inparallel lines

1) 2)

Measuring transformers

One-end Fault LocationAlgorithm Compensating for

Remote End InfeedEffect Hardware Configuration

One-end Fault LocationAlgorithm Compensating for

Remote End InfeedEffect Field Results Experienced

Installation Event Results

1 Sweden, 130 kV, 76 km P-E fault, July 1982 67.6 km67.0 km (error 0.8%)

2 USA, 138 kV, 23.3 km Five staged faul ts on paral lel Maximum error of 3%lines, Oc tober 1983 (wi thout compensat.)

3 Sp ai n, 400 k V, 135 k m P-E fau lt , Mar ch 1984 Di sp lay ed i n t he

93 to 99% of line range 93 to 99%4 It al y, 380 k V, 88 .5 k m P- E f au lt , Feb ru ar y 1984 16 % (n o er ro r)

16% of line

5 Norw ay , 45 kV, 29 .3 km P-P faul t, December 1984 77% (e rror 0.5%)77% of line

6 Fi nl an d, 1 10 k V, 130 k m P-E f au lt s, J un e 1985 Di sp lay ed i n t he78 to 90% of line range 78 to 90%

(error max 0.4%)

7 In di a, 400 kV , 23 6 k m P-E fau lt s, Dec em ber 1987 (n o er ro r)76 to 78% of line


5/13

Optimization of One-end FaultLocation

Optimization of One-end Fault Location

BA

dZL (1d)ZL

{iA}

ZAE

A

F

EB

ZB

FLd

{uA}

Aim :

improving fault location accuracy by introducingcompensation for shunt capacitances limiting influence of uncertain parameters on faultlocation accuracy to get simple formulae by applying generalized faultloop model and fault model


Symmetrical components approach appears as veryeffective technique for transposed lines and faultlocation algorithm is formulated in terms of thesecomponents (positive-, negative- and zero-sequence)

Ac

Ab

Aa

2

2

2A

1A

0A

aa1

aa1

111

3

1

V

V

V

V

V

V

)3/2exp(ja


0)(F0F0F2F2F1F1FA_P1LA_P

IaIaIaRIZdU

Generalized fault loop model:

d, RF unknown distance to fault (p.u.) and fault resistance

UA_P , IA_P fault loop voltage and current (dependent on fault type)

Z1L line impedance for the positive-sequence

IF1, IF2, IF0 symmetrical components of the ttotal fault current

aF1, aF2, aF0 weighting coefficients (dependent on fault type)


A00A22A11A_PUaUaUaU

A0

1L

0L0A22A11A_P

IZ

ZaIaIaI

AII0

1LI

0mAI0

1LI

0LI0AI22AI11A_P

IZ

ZI

Z

ZaIaIaI

a1, a2, a0 share coefficients (dependent on fault type)

Fault loop voltage and current (in terms of symmetrical components):

Fault loop voltage:

Fault loop current single line:

Fault loop current parallel lines:


F2F2F1F1F0F0F IaIaIaI

aF0, aF1, aF2 weighting coefficients (complex numbers),dependent on fault type and the assumed priority for usingparticular symmetrical components,

IF0, IF1, IF2 zero-, positive-and negative-sequencecomponents of total fault current, which are to be calculatedor estimated

Total fault current can be expressed as the weighted sumofits symmetrical components:


6/13


0F0001

2

2 RAAdAdA

1L12ZKA

A_P11L11ZKZLA

A_P10

ZLA

A_P

1A2F2A1F100

)(

I

MIaIaA

Fault location formula:

After resolving intoreal/imag parts theunknowns: d, RF aredetermined


0)(comp

F0F0

comp

F2F2

comp

F1F1F

comp

A01L

0Lsh

)1(00

comp

A2

sh

)1(22

comp

A1

sh

111L)(A_P)(

)1(

IaIaIaRI

Z

ZAaIAaIAaZdU

nn nnn

A1

th

1

'

L1)1(A1

comp

A1 )1(5.0 UAYdII

nn

A2

th

2

'

L2A2

comp

A2 )1()1(5.0 UAYdII

nn

A0

th

0

'

L0)1(A0

comp

A0 )1(5.0 UAYdII

nn

BA IAi

UAi

UFi

UBi

IBiF

IFi

IFi

IAAi

sh

L)( )1( ni

'

inAZd

th

L)1( )1(5.0

n

i

'

inAYd

sh

L)( )1()1(

ni

'

inBZd

th

L)1( )1()1(5.0

ni

'

inBYd

comp

AiI

Compensation for shunt capacitances of the line:


0 10 20 30 40 50 60

0.6

0.8

1

Dis

tancetofault(p.u.)

Fault time (ms)

No compensation

daver.

=0.7806 p.u.

0 10 20 30 40 50 60

0.6

0.8

1

Distancetofault(p.u.)

Fault time (ms)

With compensation

daver.

=0.8032 p.u.

Example: 400kV, 300km line; a-g fault, d=0.8 pu, RF=10

Due to compensation the error decreases from1.94% to 0.32%

Fault Location on Parallel Lineswith measurements at one-end

Fault Location on Parallel Lines under Availability of

Complete Measurements at One End

AB

IAB

IAA

VAA

AA

BB

BA

F

dFL

Fault Location on Parallel Lines under Availability of

Complete Measurements at One End

Traditional one-end FLs for parallel lines applythe following standard input signals:

phase voltages

phase currents from the faulted line

zero-sequence current from the healthy line(to compensate for the mutual coupling)

Limitationss of the traditional one-end FLs:

pre-fault measurements are required

remote source impedance data has to be provided


7/13

Two-end Fault Location

Two-end FaultLocation

One-terminal methods have some limitations due tonecessity of taking simplifying assumptions

Two-Terminal methods give better results but require

communications

Methods using Global Positioning Satellites (GPS)

- synchronized phasors from both ends

Methods requiring time-tagging of events

- no synchronized phasors

Low-speed communications needed for two-end faultlocation

Analyze data from two ends at a third, more convenient site

Two-end FaultLocation Synchronized Measurements

~

MUA

A B

~

MUB

GPS

FL

d,RF

RF

d[p.u.]

~

MUA

A B

~

MUB

FL

d,RF

RF

d[p.u.]

Two-end FaultLocation Unsynchronized Measurements

tA

tA=0

tB

t

tB=0

t

FLT

t=tB=0

()

(1t)

FLT DETECTION AT "A"

tFLT

FLT DETECTION AT "B"

sampling interval

TB-A

Need for phase alignment:

Two-end FaultLocation Unsynchronized Measurements Two-end FaultLocation use of incomplete measurements

Use of incomplete two-end measurements:

two-end currents and one-end voltage (2xI +1xV)

one-end current and two-end voltages (1xI +2xV)

two-end voltages (2xV)

two-end currents (2xI)


8/13

Fault location (FL) function added to current differential relay

Use of two-end synchronised measurements of three-phasecurrents and additionally providing the local three-phase voltage

SYSTEM A

A BF

DIFF

RELA

{iA}

SYSTEM B

DIFF

RELB

dA, R

FAFL

dA

ZL

(1dA)Z

L

{IB}

{vA}

{IA}

{iB}

Two-end FaultLocation use of: 2xI +1xV Two-end FaultLocation use of: 1xI +2xV

SYSTEM A

A B

F

FLCOMMUNICATION

SYSTEM B

SATUR.

dA, R

F

LAZd LA )1( Zd

jAeI

jAeV

BI

BV

pre

Immunity of fault locationto saturation of CTs at one lineside is assured by rejecting currents from saturated CTs

Three-end & Multi-end FaultLocation

Three-end FaultLocation

Use of measurements: synchronized three-phase currents from all (A, B, C) ends three-phase voltage at Fault Locatorbus A

A B

TI

A

VA

IB

FLRESULTS

CI

B

PROTECTIVERELAY'B'

PROTECTIVE

RELAY'C'

IC

IB

IC

IA

IC

IAPROTECTIVE

RELAY 'A'

FL

Solution

Fault location algorithm consists of three subroutines

(SUB_A, SUB_B, SUB_C) and the procedure for selecting

the valid subroutine

SYSTEM A

AB

T

FL

IA

SYSTEM B

VA

IB

SUB_A

FL RESULTS

CIC

dAdB

dCSYSTEM C

SUB_B

SUB_C

Selectionof faulted line section

1. Fault distance calculation assuming the fault

to be on t he AT, TB or TC segment: 3 different

results

2. Selection procedure is based on checking the

rejection conditions:

fault occurring outside the section range

calculated fault resistance has negative value

correctness of the estimated remote source

impedances

General algorithm:


9/13

FaultLocationExample

A B

TI

A

VA

IB

FL RESULTS

C

IC

PROTECTIVERELAY 'B'

PROTECTIVERELAY 'C'

FA FB

FC

IC

IB

IA

IB

IC

IA

FL

PROTECTIVERELAY 'A'

Network parameters:

Line: , (/km)

System A: ,

System B:

System C:

j0.3151)0276.0(L1 '

Z j1.0265)275.0(L0 '

Z

j3.693)+0.65125(SA1Z j6.5735)+1.159(SA0Z

SASB 2= ii ZZ

SASB 3= ii ZZ

F/km012.01 LC F/km008.00 LC

a-g fault at the section TB, dB=0.6 p.u., RFC=0.3

FaultLocationExample (1)

A BT

C

SUB_B

0 10 20 30 40 50 600

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Post fault time [ms]

Distancetofault[p.u.]

(dB)av

=0.6042

(dA)av

=1.6933

(dC)av

=0.6726

0 10 20 30 40 50 60-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

Post fault time [ms]

(RFC

)AV

= 0.6721

(RFB

)AV

=0.3232

Faultresistance[]

SUB_B is selected as valid one

Four-end Fault Location

Use of measurements: synchronized three-phase currents fromall (A, B, C, D) ends three-phase voltage at Fault Locatorbus A

SYSTEMC

SYSTEMD

Fault Location in Distribut ion(Medium Voltage) Networks

Introduction

Fault location in MV networks differs from that in HV/EHVtransmission lines

When a current of a faulty line is not directly available in theFL, certain error is introduced when assumed the current atthe substation

MV line may be multi-terminal and/or contain loops whatcreates problem in single ended fault location

In the case of MV line, there are often loads located betweenfault point and the busbar. Since the loads change and areunknown to the FL it is difficult to compensate of them

Issues forDistribution Networks

Network grounding

ungrounded networks

Petersons coil

resistance grounded

Lack of measured data for tapped loads

fault on a main or on a tap?

Unbalanced network configuration and load

Dynamic change in a network configuration

Change in conductor impedance

Multiple faults


10/13

Algorithm Structure

Estimation of the

impedance

Estimation of the

distance

Which feeder

short-circuited?

Information from

relays and/or CBs

c urrents v olt ages

impedance

distance

Digital Fault Recorder

orEMTP/ATP simulator

Fault-Loop Impedance Measurement

Z1

Z2

Zk

Zm

kC

kB

kA

k

I

I

I

I

kC

kB

kA

k

V

V

V

V

Impedance Measured at the Faulty Feeder

Phase-phase fault lo op:

Phase-ground fault loo p:

I I Ikpp kA kB

V V Vpp A B

kZ Z

ZkN

0 1

13

I I I IkN kA kB kC

Z Z0 1, Fault -loo p im pedan ces for faul t at the consi dered nod e

ZV

Ik

pp

kpp

ZV

I k I

k

ph

kph kN kN

Distance to Fault Estimation

Zpk-1 Zpk

lfk-1

Zsk-1 (1-lfk-1 )Zsk-1

Rf

k-1 k

Equivalent diagram of the cable segment with fault:

EMTP/ATP simulation with anUtility Network

Scheme of the Considered Network

Substationgrounding

HV LV

150 kV/10 kV

Zsys

RtgRg

Vsys IS

IL

VS


11/13

Scheme of Distribution Network

equivalenta equivalentb

equivalentc equivalentd equivalente

1 2 3 4

5 6 7

89

10

1112

13

14

15

16 17 18 19

20

21

grounding systemconnection

Idea of the feeder model representation: Current measured at the faulty feeder: Feeder 2.08

Distance to Fault Calculation from the Recorded Data

No File Fault type Estimated Distance

to Fault, m

1 97031400.MAT A-B GAMR-RURW - 8867 mGAMR-BJCG - 8935 m

2 97031401.MAT A-B BETR-GAMR - 8491 m


4 97031403.MAT A-G GAMR-RURW - 8780 mGAMR-BJCG - 8776 m

5 97031404.MAT A-G BETR-GAMR - 8431 m


to Fault, m


2 97031401.MAT A-B BETR-GAMR - 8491 m


4 97031403.MAT A-G GAMR-RURW - 8780 mGAMR-BJCG - 8776 m

5 97031404.MAT A-G BETR-GAMR - 8431 m

Actual faultat 8999 m

Current measured at the substation: Feeder 2.08


to Fault, m

1 97031400.MAT A-B GAMR-RURW - 8854 mGAMR-BJ CG - 8762 m

2 97031401.MAT A-B GAMR-RURW - 8745 mGAMR-BJ CG - 8755 m

3 97031402.MAT A-G GAMR-RURW - 8776 mGAMR-BJ CG - 8772 m

4 97031403.MAT A-G GAMR-RURW - 8897 mGAMR-BJ CG - 8889 m

Distance to Fault Calculation from the Recorded Data

Actual faultat 8999 m

Comparison of EMTP/ATP simulationwith recorded Stage Fault

EMTP Simulation: Comparison with Recorded Stage Fault EMTP Simulation:Comparison with Recorded Stage Fault


12/13

Conclusions

Conlusions Benefits of Fault Location

Quick elimination of permanent fault to: minimize outage time facilitate service and maintenance minimize production losses reduce cost

Pinpointing of weak spots due to temporary

fault to: assist patrol in finding excessive tree growth allow rapid arrival at the site of vandalism

Conclusions

Accurate fault location is key to improved operations andlower maintenance cost

Selection of a fault location method depends on networkconfiguration, communications, and requirements

One-terminal methods have limited accuracy

Two-terminal methods give higher accuracy

Analysis at convenient site using data from existing Pdevices

The fault location algorithmcaneasily be expandedto coverlines with three-terminals and evenmore

Fault location algorithmfor Medium Voltage Network isbased on voltage and current phasor estimation. Thealgorithm was investigated and proved on the basis ofvoltage and current data obtained from EMTP/ATPsimulations as well as recorded at DFR experiences

Fault Location on Power NetworksBook Series Power SystemsISSN 1612-1287Publisher Springer LondonDOI 10.1007/978-1-84882-886-5Copyright 2010ISBN 978-1-84882-885-8 (Print) 978-1-84882-886-5 (Online)

Fault Location On Power Networks

Fault Location on Power Lines enables readers to pinpoint thelocation of a fault on power lines following a disturbance.The nine chapters are organised according to the design ofdifferent locators. The authors have compiled detailedinformation to allow for in-depth comparison.Fault Location on Power Lines describes basic algorithmsused in fault locators, focusing on fault location on overheadtransmission lines, but also covering fault location indistribution networks. An application of artificial intelligence in this field is alsopresented, to help the reader to understand all aspects of faultlocation on overhead lines, including both the design andapplication standpoints. Professional engineers, researchers, and postgraduate andundergraduate students will find Fault Location on PowerLines a valuable resource, which enables them to reproducecomplete algorithms of digital fault locators in their basicforms.


13/13

Tableof Contents

1. Fault Location - Basic Concepts and Characteristic ofMethods

2. Network Configurations and Models3. Power-line Faults - Models and Analysis4. Signal Processing for Fault Location5. Measurement Chains of Fault Locators6. One-end Impedance-based Fault-location Algorithms7. Two-end and Multi-end Fault-location Algorithms8. Fault Location in Distribution Networks9. Artificial Intelligence Application

References (352)

Fault Location Principles

Documents

Transcript of Fault Location Principles