distribution automation smart feeders in a smart grid ... · pdf file01.09.2011 ·...

Distribution Automation –Smart Feeders in a Smart Grid World

Automatic Feeder

SwitchingBob Uluski

Switching

© 2010 Quanta Technology LLC

Introduction

• Distribution Feeder Automation is the monitoring and control of devices located out on the feeders themselves– Line reclosers

– Load break switches

– Sectionalizers

– Capacitor banks

– Line regulators


Main Feeder Automation Applications

• Automated Line Switching (ALS)

• Volt/VAR Control (VVC)• Volt/VAR Control (VVC) (Discussed in next Seminar Module)


Automatic Line Switching:h lThe Value proposition

• Reduce system SAIDI & SAIFI significantlyReduce system SAIDI & SAIFI significantly

• Accommodate and take advantage of distributed energy resourcesdistributed energy resources

• Optimal network reconfiguration– Reduce peak loading and total technical losses via load balancing


Primary ALS Application – “FLISR”

• Fault Location, Isolation, and Service Restoration

• Use of automated feeder switching to:– Detect feeder faults

– Determine the fault location (b i h )(between 2 switches)

– Isolate the faulted section of the feeder (between 2 feeder switches)feeder (between 2 feeder switches)

– Restore service to “healthy” portions of the feeder


Nature of the Problem• When a permanent fault occurs customers on “healthy”• When a permanent fault occurs, customers on healthy

sections of the feeder may experience a lengthy outage

FAULTOCCURS

CustomerReportsOutage

FieldCrews

On-Scene

FaultLocated

POWER RESTOREDTO CUSTOMERS ON HEALTHY SECTIONS

OF FEEDER

FeederBack toNormal

Travel TimeFault Investigation

& Patrol TimeTime to Perform

Manual Switching Repair Time

5 – 10 minutes

15 – 30 minutes

15 – 20 minutes

10- 15 minutes

45 – 75 minutes

• FLISR provides the means to restore service to some customers before field crews arrive on the scene


Time Line Without and With FLISRF d

FaultOccurs


Travel Time

FaultLocated

Fault Investigation& Patrol Time

Time to PerformManual Switching Repair Time

FeederBack toNormal

Without

POWER RESTOREDTO CUSTOMERS ON

5 – 10 minutes

15 – 20 minutes

10 - 15 minutes

45 75

15 – 30 minutes

1 - 4 Hours

FLISR

Field

TO CUSTOMERS ON HEALTHY SECTIONS

OF FEEDER

45 – 75 minutes

Feeder

POWER RESTOREDTO CUSTOMERS ON HEALTHY SECTIONS

OF FEEDERField

CrewsOn-Scene

FAULTOCCURS

FeederBack toNormal

Travel Time Repair TimePatrolTime


With FLISR

15 – 30 minutes

1 - 4 Hours

1 to 5

5 - 10 minutes

5 – 10 minutes

FLISR


1 to 5 minutes

Terminology – Sample Feeder

Substation #2

5

4

Substation #1 Substation #3

1 2

3

6 7 8


Terminology – Automated Switches

Substation #2Normally

• All switches must be electrically operable

5

yclosed (line)

switch• Can be load break or fault interruptingdevices

4

Normally open (tie)


1 2

3

6 7 8

open (tie) switch


Terminology – Fault Detector (FD)

Substation #2

5 FD installed on every switchFD indicates whether fault

current has passed through4

current has passed through the switch

No


1 2

3

6 7 8

Fault

Fault Current

NoFaultFaultFault


Terminology – Feeder Sections

Substation #2Section: a portion of the

5

Section: a portion of the feeder between two automated switches

4

Faulted “Healthy”


1 2

3

6 7 8

FaultFeeder Section

Feeder Section


Terminology – Sources

Substation #2“Strong” Source

• Alternate sources required f i DA b fit

5

Strong Source – Capable of

carrying a significant part of

for maximum DA benefit

• Strong or Weak Source

4feeder 1 if necessary

“Weak” Source – Capable of

carrying only a small part of


1 2

3

6 7 8

small part of feeder 1


FLISR Operation – Normal State

Substation #2

5

4


1 2

3

6 7 8


FLISR Operation – A Fault Occurs

Substation #2 Permanent fault occurs in section surrounded by switches 2, 3 and 6

5

y ,

FDs at switches 1 and 2 detect the fault

FLISR stores value of load through each switch just prior to the fault

4

j p(usually a fifteen minute average).

FLISR logic does not yet open/close any switches


1 2

3

6 7 8

No Fault

Fault Current Fault

Fault Fault No Fault


CB Trips – Feeder Deenergized

Substation #2 Circuit breaker trips

5Entire circuit de-energized (dotted line)

FDs at switches 1 and 2 i i k d

4remain picked up

Still no FLISR control actions


1 2

3

6 7 8

No Fault

Fault



CB Recloses – Fault Still There

Substation #2FDs at switches 1 and 2

5

FDs at switches 1 and 2 remain picked up

Still no FLISR control actions

4actions


1 2

3

6 7 8

No Fault

Fault Current Fault



CB Trips Again – Feeder Deenergized

Substation #2

Circuit breaker trips and locks out

Entire circuit de-energized5

Entire circuit de energized (dotted line)

FDs at switches 1 and 2 remain picked up

4

p p

FLISR open/close logic triggered


1 2

3

6 7 8

No Fault

Fault



FLISR Step 1 – Identify Faulted Section

Substation #2 FDs 1 & 2 saw a fault

5FDs 3 and 6 did not see the fault

Fault must be in section b i h 2 3 & 6

4between switches 2,3, & 6

Faulted


1 2

3

6 7 8

No Fault

Fault

Feeder Section



FLISR Step 2 – Isolate Faulted Section

Substation #2 Automatically open switches 2 3 & 6

5

switches 2,3, & 6

4

Faulted


1 2

3

6 7 8

No Fault

Fault

Feeder Section



FLISR Step 3 – Restore “Upstream” Section

Substation #2 FLISR closes CB“Upstream” =

5No need to check load –we know CB can carry the first section

FD i h # 1

between substation and faulted section

4FD at switch # 1 resets

Faulted


1 2

3

6 7 8

No Fault

Fault

Feeder Section

Fault No Fault


FLISR Step 4 – Restore “Downstream” Load(This is the tricky part)

Substation #2 Substation #2 is a “strong” source – it can carry“Downstream”

(This is the tricky part)

5

source it can carry additional load

Close switch 4 to pick up part of faulted feeder

= between faulted section

and end of feeder

4

part of faulted feederfeeder


1 2

3

6 7 8

No Fault

Fault

Fault No Fault


FLISR Step 5 – Restore “Downstream” Load (continued)( )

Substation #2 Substation #3 is a “weak” source – can’t carry additional

5

yload at this time

Switch 7 remains open

Feeder section between 4 switches 6 and 7 remain de-

energized

End of FLISR operation


1 2

3

6 7 8

No Fault

Fault

Fault No Fault


FLISR Benefits

• Reliability Improvement ‐ Significant portion of customers restored quickly (1 minute or less, versus 45 75 minutes without FLISR)versus 45 – 75 minutes without FLISR)

SAIDI Improvement Versus Baseline

52%52%50%60%

SAIFI Improvement Versus Baseline

42%42%40%

50%

33%

0%10%20%30%40%50%

Conventional recloser 1 load break w ith 1 Intellirupter w ith

18%

0%

10%

20%

30%

40%

Conventional recloser 1 load break w ith 1 Intellirupter w ithDA With Load Break

DA With DA With DA With Intelliteam II Intelliteam III Intelliteam II Intelliteam III

SAIFI(MI) Improvement Versus Baseline

21%30%

CAIDI Improvement Versus Baseline

17.9%17.9%18.3%20.0%

Break Switches

Reclosers Load Break Switches

Reclosers

-25%

6%

-30%-20%-10%

0%10%20%

Conventional recloser 1 load break w ith 1 Intellirupter w ith

9%9%

10.0%

12.0%14.0%

16.0%18.0%

Conventional recloser 1 load break w ith 1 Intellirupter w ithDA With DA With DA With DA With


Conventional recloser 1 load break w ithIntelliteam II

1 Intellirupter w ithIntelliteam III

Conventional recloser 1 load break w ithIntelliteam II

1 Intellirupter w ithIntelliteam IIILoad

Break Switches

DA With Reclosers

DA With Load Break

Switches

DA With Reclosers

Estimating Reliability Improvement Benefits From FLISRSome Useful Formulas

Note: These formulas assume that customers are evenly distributed over the length of the feeder, faults are equally likely to occur anywhere AND there is a suitable alternative source that is located downstream of the switch

• % customer minute (CM) & customers interrupted (CI) savings with DA– Assumes all DA switches have a suitable “downstream” source

– No reclosers initially:• % Improvement = (NSW) / (NSW + 1) * 100

– Where NSW = # of normally closed DA switches– Example: if NSW = 3, % improvement = 3/(3+1)*100 = 75%

Have reclosers to start with:– Have reclosers to start with:• % Improvement = 0.5 * (NSW) / (NSW + 1) * 100

– Example: If NSW = 3, % improvement = 0.5*3/(3+1)*100 = 37.5%


Other FLISR Benefits

• Labor Savings Less fault investigation and patrol• Labor Savings – Less fault investigation and patrol time because fault location is narrowed down considerablyconsiderably

• Reduction in Unserved Energy – Get some metersReduction in Unserved Energy Get some meters turning sooner


Feeder Load Balancingb d k d d• Objective: Reduce peak demand on

feeders/substations by periodically shifting load between connected feeders to achieve better balancebetween connected feeders to achieve better balance

• Must have significant diversity between feeders

5

6

7

0

1

2

3

4

01 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27


• “Make before break” to avoid momentary outages

Load Balancing – Normal Configuration

Substation #2

Sub #1 at Peak Load



Load Balancing – Sub#1 At Peak Load

Substation #2

Sub #1 at Peak Load


Load TransferredTo Sub #3


Load Balancing – Sub#3 At Peak Load

Substation #2

L d

Sub #3 at Peak Load


Load TransferredTo Sub #1


Load Balancing ‐ Benefits

• Reduction of Peak Demand on individual substation– Defer capacity addition

– Reduce individual substation demand chargesReduce individual substation demand charges

• Reduction of Electrical Losses• Reduction of Electrical Losses– Total losses with balanced load < Total losses with one heavily loaded feeder and one lightlywith one heavily loaded feeder and one lightly loaded feeder


Beyond Load Balancing –l k fOptimal Network Reconfiguration

• Objectives :l f d l ( f f l l h d )– Minimize energy losses on feeder lines (time frame of multiple hours or days)

– Balance loads among phases, feeders, substations, transformers– Plan outages for equipment or feeder section maintenance

• Seasonal OptimizationSeasonal Optimization– Goal => optimize topology for steady state operations

• Minimal power and energy losses• Maximum reliability• Best load balance• Best load balance• Best voltage profiles

• Near Real‐Time Optimization– Normal operations => Loss Optimization Mode– Emergency operations => Voltage Reduction Mode– Peak load shaving => MW Reduction Mode


Cold Load Pickup

• Objective: Reduce the time to restore loadObjective: Reduce the time to restore load following extended outage for which cold load conditions applyconditions apply

R i i i i• Restore service a section at a time via remote control


Cold Load Pickup Example

Substation #2




Substation #2


Open all switches via

remote control



Substation #2

Restore one section

t ti


at a time


Cold Load Pickup Benefits

• Reliability Improvement Benefit: Faster overallReliability Improvement Benefit: Faster overall service restoration

• Labor Savings: Fewer manual switching i i i l l iactivities, less travel time


Emergency Load Shedding

• Some utilities are limited in amount of load shedding that is possible due to presenceshedding that is possible due to presence of critical (non‐interruptible) loads

• Automated switches can be used to shed l d h f d ith t i tiload on such feeders without impacting non‐interruptibles


Load Shedding Using Line SwitchesSubstation #2

Non-Interruptible


Interruptible Customer in this section

Substation #3

Open This Switch


Use of Faulted Circuit Indicators

Remotely monitored faulted i it i di t i t i

4

circuit indicator can assist in pinpointing fault location

Substation #1

1 2

3

6 7


Benefits of FCI MonitoringBenefits of FCI MonitoringFault Prediction by OMS/AMI


Benefits of FCI MonitoringBenefits of FCI MonitoringFault Prediction by OMS/AMI & FCI


Risk – Section “Net Load” Problem

Substation #2

With DG running, pre-fault load through switch 6 is 1 MW

However, total load in this section is 3 MW (DG supplying

5

section is 3 MW (DG supplying 2MW)

FLISR may use prefault load thru switch 6 to make

d i hi d i i4

2MW

downstream switching decision

This could overload Substation 3


1 2

3

6 7 8

Prefault Load

1MW


FLISR with Large DG or Distribution NetworksDistribution Networks

Substation #2 FLISR fault detectors

5might predict fault

beyond switch number 6, and….

4


1 2

3

6 7 8

No Fault

Fault Current Fault Fault

Current

Fault Fault Fault



Substation #2 Trip switch 6, ultimately

5resulting in…

4


1 2

3

6 7 8

No Fault

Fault Current Fault

Fault Fault Fault



Substation #2 Entire feeder outage

5

4


1 2

3

6 7 8

No Fault

Fault

Fault Fault Fault



Substation #2 With directional fault

5indicators, FLISR would locate the fault properly,

resulting in

4


1 2

3

6 7 8

No Fault

Fault Current Fault Fault

Current

Fault Fault Fault



Substation #2 Power restored to as

5many customers as

possible!

4


1 2

3

6 7 8

No Fault

Fault

Fault Fault Fault


Use of Line Reclosers for FLISR

Substation #2

5

4

Line Recloser


1 2

3

6 7 8

No Fault Fault

Fault Fault Fault


Line Recloser Trips –Portion of Feeder DeenergizedPortion of Feeder Deenergized

Substation #2 If line reclosers are used instead of load break

5

instead of load break switches, only a portion of the feeder is interrupted, but….

4

Line Recloser


1 2

3

6 7 8

No Fault Fault

Fault Fault Fault


FLISR With Line ReclosersFLISR With Line Reclosers

• May have difficulty10

Current in Amperes: x 100 at 12.47 kV.

A1A1May have difficulty coordinating several reclosers in series

1

A2A2

A3A3A3A3reclosers in series– Trip faster than upstream switches 0.1

Tim

e in

Sec

onds

A4A4

upstream switches

– Trip slower than downstream switches 0 01do s ea s c es

– Trip slower than downstream fuses

PLOTTING VOLTAGE:

BY:

DATE:

NO:

12.47 kV

8-9-2005

10

0.015 10 100


Line Recloser Trips –Portion of Feeder DeenergizedPortion of Feeder Deenergized

Substation #2 May have difficulty coordinating a series of line

5

coordinating a series of line reclosers, resulting in….

4

Line Recloser


1 2

3

6 7 8

No Fault Fault

Fault Fault Fault


Line Recloser Trips –Coordination of Reclosers in SeriesCoordination of Reclosers in Series

Substation #2 Overtripping!

5

4

Line Recloser


1 2

3

6 7 8

No Fault Fault

Fault Fault Fault


This switch must operate slower than downstream fuse, but faster than upstream recloser

Using Directional Blocking Scheme

Substation #2 Controller at “6” detects fault and fault direction

5

fault and fault direction

4

Line Recloser


1 2

3

6 7 8

No Fault Fault

Fault Fault Fault

© 2010 Quanta Technology LLCRadio Radio

Using Directional Blocking SchemeUsing Directional Blocking Scheme

Substation #2 Controller at “6” detects sends blocking signal to “2”

5

sends blocking signal to 2

4

Line Recloser


1 2

3

6 7 8

No Fault Fault

Fault Fault Fault


Please don’t trip!

Using Directional Blocking SchemeUsing Directional Blocking Scheme

Substation #2 Only the recloser at switch 6 trips

5

6 trips

4

Line Recloser


1 2

3

6 7 8

No Fault Fault

Fault Fault Fault


Please don’t trip!

System Components

Distribution SCADA system

Remote controlled feeder switches– Remote controlled feeder switches

– Feeder RTU or Controller

– Fault detectorsFault detectors

– Two‐way communication facilities

( Sp.SpectrumMAS


Distribution SCADA System

• Supervisory Control and Data Acquisition (SCADA)• Supervisory Control and Data Acquisition (SCADA) system:

– Minimum requirement: Allow distribution system operator to monitor and oversee operation of the feeder automation facilitiesfeeder automation facilities

– May also perform the bulk of the feeder automation ( l d h l hprocessing (centralized scheme only – more on this in

Module 5)


Feeder Automation SwitchesAll i h b l i ll• All switches must be electrically operable

• Can be load break or faultCan be load break or fault interrupting devices

• Can be overhead or underground( )Load Break Switch (padmount) switchesLoad Break Switch.

Photo courtesy of Bridges Electric

Kyle® Type NOVA™ RecloserPad Mounted Switch


(Fault Interrupting Device).Photo courtesy of Cooper Power Systems

ABB OVR Recloser

Retrofit Motor Operators

may be able to save money by retrofitting a motor operator

Retrofit PadmountRetrofit Pole Mounted Switch

g pon an existing manual switch

Retrofit Motor-Operator

Retrofit Padmount. Photo Courtesy of S&C Electric

Co

Retrofit Pole Mounted SwitchPhoto Courtesy of Cleaveland Price, Inc


DA Switch Vendors

Vendor Name W ebsite RecloserLoad Break

bb / d t/ /9AAC

Current InterruptingPole

Mount PadmountRe trofit

KitFault

Interrupting

ABBwww.abb.com/produc t/us /9AAC720078.aspx X X

Cooper Power Systems www.cooperpower.com X X X XS&C Electric Company www.sandc .com X X X XJoslyn Hi Voltage www.joslynhivoltage.com X X X

X

G&W www.gwelec .com X X X XBridges Electric www.bridgeselec tric.com XCleaveland Price www.cleavelandprice.com X

Turner Electric Co. LLChttp://www.turnerswitch.com/installation/operators .html X

Siemens

pFederal Pac ific http://www.federalpac ific .com/ X X

NOJA Power (Australia) www.nojapower.com.au X


Feeder RTU or Local Controller

• Acquires data from FD or local sensors (or can serve as FD!)

• Executes control commands

S&C 5800

• May perform bulk of feeder automation processing (distributed architecture) SEL 451

S&C( )

• Provides interface to communication facilities

S&C UIM

DAQ Polaris Pole Top

Cooper Form 6 Pole-Top

RTU Controller


Fault Detector (FD)

• Determines that a fault has occurred “downstream” (further from the substation)

• Basic Requirementsq– Must be able to identify and “capture” the fault information before fault is cleared (in a few

l )cycles)

– Must be able to detect all kinds of faults

• Phase faults• Phase faults

• Ground faults – current may be less than normal load – must use “residual” current


Two Ways to Accomplish FD

• Faulted Circuit• Faulted Circuit Indicator (FCI)

• Current/Voltage Sensor


Faulted Circuit IndicatorsF lt d Ci it I di t (FCI)

Photo Courtesy of Horstmann/Power Delivery Products

Faulted Circuit Indicator (FCI)Close a contact when pre-established ampere threshold is exceeded

– Clamp on style

– Current inrush restraint

exceeded

– Must work in either direction

– Reset conditions

• Time

• Restoration of voltage or current

– Local indicator visible from ground level

– Output signal to feeder RTUEdison Controls FD

p g

• Radio signal

• Fiber optic/metallic cable


Radio Transmitter Style Fault Indicator


Current & Voltage Sensors

– Perform basic current & voltage measurementProvide measurement to a separate device– Provide measurement to a separate device for analysis (“does measurement exceed threshold?”)

– Doubles as data acquisition unit for feeder ( )monitoring (SCADA) system

– Voltage sensor may also serve as a power source for DA feeder equipment

– Basic requirementsFisher Pierce Line P t SBasic requirements

• Accuracy at least + or ‐ 3%• Suitable for measuring fault current (should not saturate)C b t d l it( ) i t l t f

Post Sensor

• Can be standalone unit(s) or integral part of switch itself

Lindsey CurrentSensor


Current & Voltage Sensors

Can be an integral part of the DA switch

Current-Voltage Sensor

“S d M t ” b S&C El t i“ScadaMate” by S&C Electric


Bidirectional Fault Indicatorsid f l d i d• Provide fault detection and

direction of fault• Useful Required for

– Systems with DGSystems with DG– closed loop, paralleled or

networked distribution systems

Power Delivery Productshttp://www.powerdeliveryproducts.com/directionalfaultindicator.htmQuanta Technology has seen a

growing number of utilities using


growing number of utilities using directional overcurrent relays

for this purpose

Communication Facilities

• System requires reliable 2‐way communication facilities to feeder locations

• Common approaches:– Licensed UHF MAS radio

– Unlicensed Spread spectrum radio

– Cellular telephone & other commercial services

– Power line carrier

– WiMAXMore details on DA Communications laterMore details on DA Communications later


Control Panels

• Open/Close Pushbuttons

h ( / l d) d• Switch status (open/closed) indicators

• Alarm indicators

• Local/remote switch

• Operations counterOperations counter– Electrical operations

Mechanical operations?– Mechanical operations?


Switch Power Supply (SPS)

• Key Point:Most switch operations run off the SPS with the line dead!– Should specify a required number ofShould specify a required number of

operations with the power off (biggest power drain is radio transmitter)

– As an absolute minimum must be able toAs an absolute minimum, must be able to open switch and close switch once without recharging (with the line dead)

• Power SourcePower Source– Voltage sensor may provide all the power

needed by the switch, controller, and radio

If not use dedicated voltage transformer or– If not use dedicated voltage transformer or connect to local secondary circuit


Categories for Automated Line Switching

• Approaches categorized on how control isApproaches categorized on how control is performed:– SupervisedSupervised

– Semi‐Automatic

– Fully Automatic


Feeder Automation CategoriesS i d• Supervised

• No automatic control – “person in the loop”• System delivers information (recommendations) to Dispatcher• Dispatcher initiates remote control actions

– Pro’s• Simpler than fully automaticSimpler than fully automatic• Good “starter” approach until confidence is built up

– Con’s• Takes longer to restore service (3 5 minutes)• Takes longer to restore service (3 – 5 minutes)

– Communication time (both ways)– Dispatcher decision time

• Difficult for dispatcher to manage many switches duringDifficult for dispatcher to manage many switches during emergencies involving multiple disturbances


Feeder Automation Categories• SEMI‐AUTOMATIC

– Mix of automatic and supervised control– ExampleExample

– DA system automatically isolates fault and performs “upstream” restoration (the “easy” part)

– Dispatcher supervises “downstream” restoration activities based on DA system recommendations (the “hard” part)on DA system recommendations (the hard part)

– Pro’s• Simpler than fully automatic• Natural progression from supervised approach• Where many utilities end up• “Upstream” customers restored in less than 1 minute

– Con’sCon s• Takes longer to restore “downstream” service (3 – 5 minutes)• Difficult for dispatcher to manage many switches during emergencies involving multiple disturbances

• Not supported by some DA vendors


• Not supported by some DA vendors

Feeder Automation Categories– FULLY AUTOMATIC

– All fault isolation and restoration activities performed automatically

Current state of the art– Current state of the art

– No dispatcher intervention

• Pro’s

– Possible to restore all service in less than one minute

– Less burden on Dispatcher to manage the switching activities

• Con’s

– Most complex approach

– Acceptance difficultiesAcceptance difficulties

» Ranges from “Why not?” to “Over my dead body”


Practical Matters to ConsiderPractical Matters to Consider


• Tradeoff: “Permanent” vs. “Momentary” Outages

• Definitions:

Permanent: Duration > threshold– Permanent: Duration > threshold

– Momentary: Duration < threshold

• Use of FLISR will:

Improve permanent outage statistics– Improve permanent outage statistics

• SAIDI, SAIFI, CAIDI

– Make momentary outage statistics worse

• MAIFI• MAIFI

• Most utilities are willing to accept this tradeoff!


• Limitations on Transferring Load to Adjacent Feedersj– It is often difficult to transfer all the “healthy” load

to adjacent feeders without causing overloads and/or voltage problems/ g p

– Especially true during peak load period

– May need to split load being transferred to alternative sources

– May require additional automated switches and addition tie points to accomplish FLISR objectives p p jat certain times of the day

– Key Point: Must have backup source to get incremental reliability improvement beyondincremental reliability improvement beyond simple recloser!

– Can use DG, Energy Storage, and other Distributed f d


Energy resources for downstream restoration (More later in DG Section – Farid)……

Estimating Reliability Improvement BenefitsA h U f l F l (C i d)Another Useful Formula (Continued)

• reduction in reliability improvement if some switches do not have a suitable downstream source

% reduction in reliability Improvement = 100 *– % reduction in reliability Improvement = 100

• Where: NDSS = # of DA switches that do not have a suitable downstream source

• Example: with total of 4 normally closed DA switches (NSW = 4), what is the reduction in benefits if 2 of these switches does not have a suitable downstream source?

100 * (1 + 2) / (4+1)2 = 300 / 25 = 12% reduction in savings


Reliability Improvement versus Number of Switches ‐Diminishing Returns

Additional reliability improvement benefit declines dramatically as more switches are added

Reliability Improvement

90100

a e added

5060708090

ovem

ent

No reclosers

1020304050

% Im

pro

Reclosers

00 1 2 3 4 5 6 7

Number of DA Switches


Reliability Improvement versus Number of SwitchesUsually 2 ½ switches is best for reliability improvement BFTB – more

switches added due to load transfer constraints, critical customers, and unusual feeder configuration

Reliability Improvement

90100

5060708090

ovem

ent

No reclosers

1020304050

% Im

pro

Reclosers

00 1 2 3 4 5 6 7

Number of DA Switches


Importance of Switch Placement• Predicted reliability improvement varies widely withPredicted reliability improvement varies widely with

switch placement strategy

• KEY POINT!: Splitting customer count into equal physicalKEY POINT!: Splitting customer count into equal physical parts is only best when customers and fault exposure are evenly distributed across the feeder (rarely the case!)

• Hypothetical Case – all customers concentrated in one spot

• If Length “A” is really small, and

Backup source

N/OPrimary source

• the backup sources is strong (able to carry the entire sample feeder)

• SAIDI and SAIFI for concentrated load could be nearly reduced to

All Load Concentrated

here

load could be nearly reduced to zero!.....


Importance of Switch Placement• Predicted reliability improvement varies widely withPredicted reliability improvement varies widely with

switch placement strategy

• KEY POINT!: Splitting customer count into equal physicalKEY POINT!: Splitting customer count into equal physical parts is only best when customers and fault exposure are evenly distributed across the feeder (rarely the case!)

• Hypothetical Case – all customers concentrated in one spot

If L th “A” i ll ll d

Backup source

N/OPrimary source

• If Length “A” is really small, and • the backup sources is strong

(able to carry the entire sample feeder)

• SAIDI and SAIFI for concentratedAll Load

Concentrated here

SAIDI and SAIFI for concentrated load could be nearly reduced to zero!.......

Hypothetical case is not realistic, but


yp ,shows that its beneficial to surround concentrated load with DA switches

Importance of Switch Placement• Switch placement analysis is not trivial – should use

engineering analysis tool!

• Sample variation between equal physical division and optimal placement :– SAIDI ‐ 22%– SAIDI ‐ 22%– SAIFI ‐ 31%– MAIFI ‐ 23%

• Small change in placement (a few hundred feet) produced a 5% change in SAIDI!


Single Phase Trippingg pp g• Some newer line reclosers support single phase tripping

• Can perform FLISR operation on only the faulted phase – avoid service interruption on unfaulted phases (SAIDI, SAIFI benefit)p ( , )

• Issues:– Adverse impact on 3 phase loadsR i i t h t i d l l d– Remaining two phases may trip ground relay on load imbalance

• Typical solution:– Trip and reclose single faulted phase for temporary fault– Trip and lockout all three phases for permanent fault


Faults on Fused Laterals

• No incremental benefit for permanentfaults on fused lateralsfaults on fused laterals– Must take this into account when computing potential benefitspotential benefits

• For temporary faults:– Can apply fuse saving

– New S&C “TripSaver” Dropout Recloser


First Segment Fault Detection• Always an issue!Always an issue!

• System requires a “lockout for fault”• System requires a “lockout for fault” signal from the substation to trigger feeder switching activitiesfeeder switching activities– Fault has occurred– Feeder protection has completed itsFeeder protection has completed its automatic reclosing cycle

• Works best if a protective relay IED is available in the substation and can be


interfaced to the FLISR system

Non‐Fault vs Fault TrippingS b bl di i i h b “• System must be able to distinguish between “non fault” and “fault” tripping of the substation circuit breaker/recloser

• “Fault” Tripping– A feeder fault has occurred or supply has been lost due to a

t i i b t ti f lttransmission substation fault

– FLISR should attempt to restore service

• “Non‐Fault” Trippingpp g– Substation CB tripped for reasons other than a feeder fault

• Manual operation by switching personnel or supervisory

control from the control center

• Underfrequency/undervoltage load shedding

– FLISR should not attempt to restore service


FLISR should not attempt to restore service

Impact on Feeder Protection

Substation #2 Entire feeder must

5remain protected following FLISR operation!.....

4


1 2

3

6 7 8



Substation #2 So if source at

5substation 1 is lost.....

4


1 2

3

6 7 8



Substation #2 FLISR May

5reconfigure feeders as shown…..

4


1 2

3

6 7 8



Substation #2

Feeder relays at “5” must be able to reach all the way to switch 1

5

y

Note: relay IEDs can switch between setting groups but most DA/DMS

4groups but most DA/DMS systems don’t auto switch the settings


1 2

3

6 7 8


Safety Issues

• Safety for workers and general public must not be compromised!!!

• Operating practices and procedures must be reviewed and modified if necessary to address presence of automatic switchgearaddress presence of automatic switchgear

• Safety related recommendations:Requirement for “visible gap”– Requirement for “visible gap”

– No automatic closures after two minutes have elapsed following the initial faultp g

– System disabled during maintenance (“live line”) work


distribution automation smart feeders in a smart grid ... · pdf file01.09.2011 ·...

Documents