IntelliGrid Architecture

 

 

Home

Parent Menu

Overview of Functions
Market Operations
Transmission Operations
Distribution Operations
HV Generation
Distributed Resources
Customer Services
List of IT Use Cases
List of Functions

Same Level Menu

Substation Data Acqu
Automated Control Base
Auto Restoration
Contngcy Analysis Base
Contngcy Analys Future
Emergency Control Base
Inter-Area Oscill Damp
Self-Healing Grid
Synchro Phasor
Systemwide Volt Control
Voltage Security
List of Brief Trans Functions

Child Menu

 

Search IntelliGrid Site

Questions/Comments

Questions

 

Responses

 

 

Transmission Operations - Wide-Area Control System Advanced Auto-Restoration

Contents

Narrative

Overview

The purpose of advanced auto-restoration is to automatically restore power to un-faulted sections of a line or feeder, after a fault is isolated, in networks having complex topologies and multiple organizational boundaries.

Currently, automatic restoration of service is performed only within a restricted set of conditions and network topologies, as described in the WACS Automated Controls Baseline use case.  In the near future, it is expected that these restrictions will be removed and the automation system will be able to restore power in systems which:

·        There are multiple sources from which to restore power

·        The multiple sources may belong to different organizations

·        There are multiple possible connection points between the sources

·        It is necessary to split the de-energized load into sections because any one source cannot re-energize the whole load

The remainder of this narrative describes an example scenario illustrating these capabilities.

Initial State

As shown in Figure 1, two neighboring substations are connected in a manner to make traditional auto-restoration possible, in other words:

·        Per typical utility operation, there is breaker located in the substation connected to each feeder, provided with an automatic reclosing function.  These are labeled 1A1, 1B1, 2A1, and 2B1 in the figure, following the naming convention <substation1/2><feederA/B><switch/breaker#>.

·        Normally-closed switches are located at intervals along each feeder to permit auto-sectionalizing around a fault.  (e.g. 1B2, 1B3, 1B4).  These switches are typically of the “no-load break” variety, for economic reasons.  They can open only when there is no load on the line.  Some may be of the “load break” variety, which can open under normal current.  Usually only those devices at the head of the feeder (such as 1A1, 1B1 etc.) will be “fault interrupting” breakers capable of opening under fault current.

·        A normally-open switch is located at the end of adjacent feeders. (e.g. 1C or 2C).  This switch can be closed to share load or restore power from one feeder to the other.

·        Each breaker and switch is monitored and controlled by an Intelligent Electronic Device (IED).

·        A Substation Computer (SC) in each substation gathers information and controls the IEDs connected to its feeders.  It reports to the Operator for that utility by way of a Graphical User Interface (GUI).

In this example, the two adjacent sets of feeders can also be connected to each other (if necessary) using a number of normally-open switches (1X, 1Y and 1Z).  This interconnection is rarely performed because the two substations belong to different utilities.  In this example, the interconnection switches are owned by the utility that controls Substation 1.  However, Operator 1 must have approval from Operator 2 before closing any of these switches.

The scenario begins with each IED reporting its downstream load and switch status to the Substation Computer.  For the purposes of this example, we assume that all this information is reported to both Substation Computers.  There are two ways to do this:

·        Each IED reports its data separately to each Substation Computer

·        Each IED only reports to its “own” Substation Computer, and the Substation Computers exchange information. 

The latter case is most likely to be implemented because:

·        It reduces the number of communications connections between the two utilities, which is desirable for security reasons.

·        It reduces the bandwidth and processing power required by each IED.

Figure 1 Initial System State

From the current data reported by each IED (shown in italics with arrows), the Substation Computers can calculate the load on each individual section of the feeders.  This example assumes that the maximum capacity limit on each feeder is 100A.  Feeder 1B, in that case, is operating near capacity, while the other feeders are about 50% loaded.

Fault Detection

As shown in Figure 2, a fault occurs on feeder 1B between switches 1B2 and 1B3.  Breaker 1B1 trips and de-energizes 90A of load, including 60A that is downstream from the fault.

All IEDs on feeder 1B report to Substation Computer 1 the fault and the loss of current.  Those IEDs that saw the fault current (1B1 and 1B2) may send an estimated distance to the fault.  IED 1B1 reports that it has tripped and has started reclosure timers.  Substation Computer 1 forwards the information to Substation Computer 2, but SC2 takes no action because the fault is not in its territory.

Figure 2 Fault Detection

Auto-Sectionalization

As shown in Figure 4, the IEDs on feeder 1B take action to isolate, or auto-sectionalize, the fault.  There are two possible methods for doing so, with different communications requirements.

·        High-Speed Communication.  One possible method is that Substation Computer 1 determines which two switches (1B2 and 1B3) to open using fault direction and distance information provided by the IEDs.  This method would require fast communication between the 1B IEDs and SC1, in order to open the switches between reclosings of the breaker (measured in seconds).  It would likely also require the IEDs to provide a specialized communications service, i.e. “open the next time you see zero current”.  

·        Fault-Interruption Counting. A more robust and distributed method would be for each IED to be programmed to open its switch after a pre-configured reclosure attempt.  Each IED would open its switch under the following conditions:

o       The IED has observed fault current

o       The IED has seen the fault current drop to zero, indicating the breaker has tripped

o       These two conditions have occurred a pre-configured number of times.  The number is different for each IED on the feeder.

 Figure 3 illustrates how this occurs in the example.  No IED is permitted to open its switch between the initial fault and the first reclosure attempt, in case the fault is transient.  1B4 is permitted to open its switch between the first and second reclosure attempts, but does not do so.  Because 1B4 is downstream from the fault and has no other source of current, it does not observe the fault current and its opening conditions are therefore not met.  Similarly, 1B3 does not observe fault current and so does not open in its time window.

IED 1B2, however, has seen the same current as 1B1, and has been counting the fault interruptions.  After the third reclosure attempt, 1B2 opens its switch, isolating the fault from any source of current.  This is auto-sectionalization, and is shown as step (1) in Figure 4.  When 1B1 recloses the fourth time, it is successful, and 10A of load is restored to that section of feeder 1B.  This is step (2) in Figure 4.

 

Figure 3 Fault-Interruption Counting for Auto-Sectionalization

Figure 4 Auto-Sectionalization and Load Splitting

Isolating the Fault

The final step in auto-sectionalization, shown in step (3) of Figure 4, is to isolate the fault.  Substation Computer 1 observes that 1B3 and 1B4 have reported zero current and voltage without having reported fault current.  It therefore determines (possibly with the assistance of distance-to-fault data from 1B1 and 1B2) that the fault is between 1B2 and 1B3.  Substation Computer 1 recommends to Operator 1 that switch 1B3 be opened in order to isolate the fault.  Operator 1 confirms this operation, and SC1 sends the message to 1B3 causing it to open.

 

Load Splitting

Whichever auto-sectionalizing method is used, the fault is now isolated and auto-restoration can begin.  Substation Computer 1 reviews the data provided prior to the fault.  It calculates the loading on each segment of each feeder, as shown in Figure 4.  It determines that there is 60A of load that can be restored.

However, the “traditional” solution, to close switch 1C, will not solve the whole problem.  Feeder 1A is already loaded at 60A.  If it accepts the whole downstream load of 60A, it will be overloaded, since the example began with the assumption of 100A maximum limit per feeder.

The Substation Computer determines that it will be necessary to “split” the downstream load and re-energize it from multiple sources.  Substation Computer 1 recommends to Operator 1 that switch 1B4 be opened, receives confirmation from Operator 1, and opens the switch by sending a message to 1B4.  This is step (4) in Figure 4.

Auto-Restoration

The final steps in auto-restoration are shown in Figure 5.  Utility 1 has a policy in place that load is to be restored from Utility 1 sources whenever possible.  Therefore Substation Computer 1 recommends that switch 1C be closed, rather than, for instance, switch 1Z.  Operator 1 confirms this operation and SC1 sends the message to IED 1C, restoring 30A of service.

Substation Computer 1 recommends that switch 1Y be closed to restore the remaining un-faulted section of feeder between 1B3 and 1B4.  Operator 1 contacts Operator 2 at Utility 2, requesting permission to close switch 1Y. 

Before making this decision, Operator 2 does the following:

·        Reviews the sequence of events logs generated by SC2 showing the auto-sectionalizing sequence.

·        Confirms that Utility 1 has isolated the fault between 1B2 and 1B3.

·        Confirms from records generated by SC2 that the de-energized section between 1B3 and 1B4 previously was loaded at 30A.

·        Checks on the SC2 GUI that feeder 2A can handle the additional 30A load.

Finally, Operator 2 contacts Operator 1, giving permission to close switch 1Y.  Operator 1 confirms the operation with SC1, which sends the message to 1Y and restores the remaining 30A of service.

Figure 5 Auto-Restoration

Load Balancing

Following auto-restoration, feeder 1A is loaded at 90A and 2A is loaded at 80A, while 2B is only loaded at 50A.  Operator 2 may choose to close switch 2C in order to lighten the load on feeder 2A.

In theory, the whole system could be more efficiently loaded by also closing switch 1Z.  However, neither Substation Computer would make this recommendation because:

·        The power on the two feeders is likely incompatible due to differences in frequency, voltage, and phase angle.  Therefore, it would be necessary to open 1C before closing 1Z.

·        Opening 1C would cause a momentary outage downstream of 1B4.  Furthermore, if 1C was not a “load break” switch, it would be necessary to first break the load at 1A1, meaning that the outage would occur for all of feeder 1A.

·        Utility 1 would lose the 30A of load downstream of 1B4 to Utility 2 until the fault could be repaired.  This would be unacceptable from a business point of view.

Summary

Performing advanced auto-restoration will require the following measures beyond those required for existing auto-restoration mechanisms:

·        Real-time sharing of data between Substation Computers

·        Calculation of loads on each feeder or line section, and storing these recent historical values in the Substation Computer.

·        More advanced logic in each Substation Computer to evaluate each possible switching action, perhaps on the order of the Contingency Analysis programs currently used by EMS stations.

·        Reliable communications between neighboring operators, either by voice or by data

·        One of the following features:

o       Full breakers and protection relays on each section, or “load break” or “fault-interrupting” switches.  Utilities are unlikely to do this because of the significantly higher cost.

o       Fault-Interruption counting, as discussed in this example. Fault-interruption counting has one major drawback:  Ideally, it requires the same number of reclosures as there are switches on the feeder.  Typically, utilities do not use a high number of reclosures because:

-         It causes excessive wear on the breaker

-         It annoys the customers, who see multiple small outages within a short period of time.

Therefore it is rare to see more than two or three reclosures.   This example, with four reclosures, would be extremely rare.  This limits the granularity with which load can be restored, and increases the number of subscribers affected by an fault.

o       High-speed communications between remote IEDs and the substation computer.  In this example, it would permit the Substation Computer to immediately determine that 1B2 and 1B3 switches should open, and do so quickly, between the first and second reclosings of breaker1B1.   This is shown as an alternate scenario in the use case below.

Steps

 

#

Event

Name of Process/Activity

Description of
Process/Activity

Information Producer

Information
Receiver

Name of Info Exchanged

IntelliGrid Architecture Environments

1A

Fault

Report Fault

IEDs upstream from the fault report seeing fault current

IED 1B1, IED 1B2

Substation Computer Device  1

Fault Detected

Deterministic Rapid Response Intra-Sub

1B

Loss of current and voltage

Report Loss of Service

IEDs downstream from the fault report loss of current and voltage

IED 1B3, IED 1B4

Substation Computer Device  1

No Current Detected,

No Voltage Detected

Deterministic Rapid Response Intra-Sub

2.1

 

Initial Trip

First feeder IED (relay) trips breaker and reports the action.  Starts reclosure timer.

IED 1B1

Substation Computer Device  1

Trip

Deterministic Rapid Response Intra-Sub

2.2

Recloser timer expires

First Reclose Attempt

Upon expiry of reclosure timer, first IED recloses the breaker.  Reports the action.

IED 1B1

Substation Computer Device  1

Switch State (close)

Deterministic Rapid Response Intra-Sub

2.3

Fault

Report Fault

IEDs upstream from the fault report seeing fault current again.  This message indicates that the fault was not intermittent and that the SC should attempt to auto-sectionalize.

IED 1B1, IED 1B2

Substation Computer Device  1

Fault Detected

Deterministic Rapid Response Intra-Sub

2.4

 

2nd Trip

First IED trips breaker and reports the action.  Starts reclosure timer.  This message indicates that the SC can now attempt to open a switch for auto-sectionalization.

IED 1B1

Substation Computer Device  1

Trip

Deterministic Rapid Response Intra-Sub

2.5

 

Auto-sectionalize

Computer determines the correct switch to open based on the fact that the upstream switches reported fault current, while the downstream switches reported no current or voltage.

Directs the correct switch to open between reclosures of the breaker.

Substation Computer Device  1

IED 1B2

Switch Control (open)

Deterministic Rapid Response Intra-Sub

2.6

Reclosure timer expires

Report Upstream Power Restored

Upon expiry of the reclosure timer, first feeder IED recloses the breaker.  Reports the action. Power is now restored from the substation to switch 1B1. 

IED 1B1

Substation Computer Device  1

Switch State (close),

Current,

Voltage

Deterministic Rapid Response Intra-Sub

3.1

Logic timer expires

Request Isolation

Computer detects (using a timer) that no fault has occurred since 1B1 reclosed the breaker.  Determines that switch 1B3 is the first switch downstream from the fault and should be opened.  Requests confirmation from System Operator .

Substation Computer Device  1

System Operator  1

Request

 (open 1B1)

User Interface

3.2

 

Confirm Isolation

Tells the Substation Computer Device  it is permitted to open the first downstream switch (1B3).

System Operator  1

Substation Computer Device  1

Confirm

User Interface

3.3

 

Isolate Fault

Requests that the first downstream switch (1B3) open.

Substation Computer Device  1

IED 1B3

Switch Control (open)

Deterministic Rapid Response Intra-Sub

3.4

 

Report Isolation Complete

The IED controlling the first downstream switch reports that the switch is open.

IED 1B3

Substation Computer Device  1

Switch State (open)

Deterministic Rapid Response Intra-Sub

4.1

 

Request Load Split

Computer determines from the Current and Voltage information stored prior to the fault that service cannot be restored from a single source.  Determines which switch to operate (1B4) and requests confirmation from the System Operator .

Substation Computer Device  1

System Operator  1

Request

(open 1B4)

User Interface

4.2

 

Confirm Load Split

System Operator  confirms that the Computer may open the switch to split the load (1B4)

System Operator  1

Substation Computer Device  1

Confirm

User Interface

4.3

 

Split Load

Computer opens the switch to split the load (1B4)

Substation Computer Device  1

IED 1B4

Switch Control (open)

Deterministic Rapid Response Intra-Sub

4.4

 

Report Load Split Complete

IED (1B4) reports that the switch is open and the load is split.

IED 1B4

Substation Computer Device  1,

Substation Computer Device  2

Switch State

Deterministic Rapid Response Intra-Sub

5.1

 

Request Local Restoration

Computer determines that half the load can be restored by closing the local normally open switch (1C), and requests permission from operator.

Substation Computer Device  1

System Operator  1

Request

(close 1C)

User Interface

5.2

 

Confirm Local Restoration

System Operator  confirms that the Computer may close the normally open switch (1C)

System Operator  1

Substation Computer Device  1

Confirm

User Interface

5.3

 

Restore from Local Source

Computer closes the switch to restore power from the local source (1C).

Substation Computer Device  1

IED 1C

Switch Control (close)

Deterministic Rapid Response Intra-Sub

5.4

 

Local Restoration Complete

IED (1C) reports that the switch is closed and current is restored to half the load.

IED 1B4

Substation Computer Device  1,

Substation Computer Device  2

Switch State (close),

Current

Deterministic Rapid Response Intra-Sub

6.1

 

Request Inter-Utility Restoration

Computer determines that the other half of the load can be restored by closing the inter-utility switch (1Y), and requests permission from operator.

Substation Computer Device  1

System Operator  1

Request

(close 1Y)

User Interface

6.2

 

Request Linking Utilities

System Operator  1 verifies Computer 1’s request and forwards it to System Operator  2 at the other utility.

System Operator  1

System Operator  2

Request

(close 1Y)

User Interface

6.3

 

Confirm Linking Utilities

System Operator  2 verifies the request and gives System Operator  1 permission to proceed.

System Operator  2

System Operator  1

Confirm

User Interface

6.4

 

Confirm Inter-Utility Restoration

System Operator  confirms that the Computer may close the normally open inter-utility switch (1Y)

System Operator  1

Substation Computer Device  1

Confirm

User Interface

6.5

 

Restore from Inter-Utility Source

Computer closes the switch to restore power from the other utility source (1Y).

Substation Computer Device  1

IED 1Y

Switch Control (close)

Deterministic Rapid Response Intra-Sub

6.6

 

Inter-Utility Restoration Complete

IED (1Y) reports that the switch is closed and current is restored to the remaining load.

IED 1Y

 

Substation Computer Device  1,

Substation Computer Device  2

Switch State (close),

Current

Deterministic Rapid Response Intra-Sub


 

IntelliGrid Architecture
Copyright EPRI 2004