1/8

INTEGRATED SUBSTATION AUTOMATION ENABLE NEW STRATEGIES FOR POWER T&D

Volker Lohmann

ABB Power Automation Ltd, Baden/Switzerland

[email protected]

ABSTRACT INTRODUCTION

In view of the envisaged competitive electricity market In the increasingly competitive arena there is significant
era, utilities are examining the application of pressure on power providers for greater system
information technology (IT) as an alternative to support reliability and improvement of customer satisfaction,
corporate business strategies that focus on improving while similar emphasis is placed on cost reduction.
service, power quality and reducing cost of operation These cost reductions focus on reducing operating and
and maintenance. Key issues for the improvement of the maintenance (O&M) expenses, and minimizing
overall productivity are more and better information investments in new plants and equipment. If plant
concerning the entire system behaviour, service investments are to be made only for that which is
condition of physical assets, e.g. circuit breakers and absolutely necessary and the existing system equipment
power transformers, as well as cost efficient operation must be pushed to greater limits in order to defer capital
and maintenance management. investments.
Substation Automation (SA) in comparison with a Utility executives, on the other hand, are increasingly
remote terminal (RTU) concept means decentralisation examining automation solution alternatives to support
of computer capacity, which shall be used e.g. for corporate business strategies that focus on improving
intelligent automatic power restoration procedures after service reliability and reducing cost of operation and
faults and for powerful decentralised processing of maintenance.
service and environment condition related data.
In the past, utility engineers often struggled with the
It is suggested that intelligent electronic devices (IED) fact that too little data was available when they
shall be implemented for protection and control tasks in attempted to analyze problems within their power
substations to substitute electro-mechanical or static systems. They also did not have enough information to
devices. In a modern SA system theses IEDs provide an predict or ascertain the level of maintenance needed of
infrastructure to collect, to process and to transmit data when it would be required for the major equipment
and information, which are utilised for cost effective located within their substations. As new and higher
condition monitoring of circuit breakers, power levels of digital technologies have made its way into
transformers, instrument transformers etc. substations in terms of numerical protection devices and
This idea is illustrated by an example for power control systems, these same engineers are suffering from
transformer condition monitoring. The SA system data overload. They have more data than can be
provides a user-friendly overview for the actual processed and assimilated in the time available. Today
transformer condition. This comprises condition-related the challenge is to automatically convert data to
parameters like: service current, temperatures, and knowledge, which frees manpower to implement
corrective or preventive maintenance.
quality of insulation oil, partial discharge measurements
and loads.
The integration of the various SA systems in a high WORKING PLANTS HARDER
performance communication network allows system
wide adaptive protection and real time automatic power Various aspects of the utilities needs have been revealed
restoration procedures in the Cigré 99 London Symposium on the subject:
“Working Plant and Systems harder”. Enhancing the
management and performance of plant and power
Keywords: systems is being discussed widely at many international
conferences. The overall conclusion perceived is that
Substation automation, condition monitoring, on- line there are a lot of new technologies available, which will
monitoring of transformers, Internet and browser help planners and operators to find new solutions to
technology, adaptive protection, wide area protection, maximise the use of the power systems and adapt to the
fast changing environment.
 2/8

There are two aspect where Substation Automation (SA) protective system isolates the assets from the
can contribute potential benefits to achieve more power: electric grid, and by means of network redundancy
(Figure 1) avoids a power outage at all, or leaves a limited
1. Better information, which result in higher area without power. With respect to the condition
productivity of assets, however, this temporary disturbance most
2. Intelligent automation which assures higher likely caused accelerated wear.
availability. Condition monitoring mainly addresses the wear and
ageing caused by normal or temporarily abnormal
working conditions. First, in that they support the
evaluation of the actual condition of assets, and second,
in that they might explicitly support the prediction of
the further evolution of a detected problem, and the
better Substation intelligent probability of breakdown. However, many of today's
provides Substation enables condition monitoring systems leave the assessment of
information Automation automation the future to the human's interpretation based on his
Automation conclusions drawn from the current status. Whichever,
R I
even if a utility decides, e.g., based on risk management
e m
considerations, to let a worn out asset in operation until
s p
it breaks, the breakdown will be a planned one, and so
u a
will the repair action be. Hence, the power interruption
l Efficient c
will most likely be rather short and the problems posed
t communication t
by the interruption alleviated as good as possible.
Apart from monitoring the condition of primary
equipment and thereby attempting to proactively
prevent power interruptions, an elaborate post fault
higher higher analysis supported by monitoring systems is equally
productivity leads to More power leads to availability important. It has been observed that a large proportion
of major blackouts of electric power systems is caused
by protective system failures. These failures are
Figure 1: The potential benefits of Substation Automation
generally hidden and only exposed during the rare
The prerequisite, however, is efficient communication occasion of system disturbances. According to utility
for easy access from remote to primary equipment opinion derived from a questionnaire over 60% of these
condition related information and fast inter-station failures are based on wrong protection settings,
protection calibration, or protection maintenance. It is
communication. (Figure 2)
therefore important to capture as much details as
possible during a system disturbance and have access to
as much protection relevant data as possible during the
Protection / Operation / Planning / Asset entire analysis. The conceivable subsequent settings
Engineering Maintenance Centre Management refinement phase is a measure to prevent the same
EMS / SCADA interruption from happening again, or, at least, minimise
Centre Centre its impact on the power distribution
Centre 1

EMS / SCADA

Centre 2 Real time Corporate

System wide Communication Backbone

Master Clock Network Network

Intersubstation

Substation 1 Substation 2 Substation 3 real time Substation n

communication

1. The breakdown of a utility asset through normal

Figure 2: Efficient communication network
wear and ageing under working conditions.
2. The breakdown of an asset being effected by an
BETTER INFORMATION
external event (system disturbance), such as a tree
falling on an overhead line that led to a permanent
For once neglecting outages as a result of wrong human abnormal working condition.
operation, there are basically three reasons for power 3. A temporary system disturbance where either the
interruptions:
external influence disappears ("self-healing"), or a
Data Acquisition

With computing power making its way into the primary

equipment, more and more equipment internal data can
be made available to the outside at virtually no cost.
Interfaces to acquire such internal data were previously
not provided for cost reasons. Data that will be
accessible includes, but is not restricted to:
• switching counters,
• thermal information,
• quality of isolation media,
• entire timing curves of switching operations,
• switching currents,
• manufacturing data,
• original value of key performance criteria.
 3/8
centred maintenance (RCM) policy. This calls for
differentiation between the following four types of
This kind of data can be the source of valuable maintenance policy:
condition information and exploited for building • Predictive or condition based maintenance, i.e. to
condition monitoring systems for those assets that monitor if something is going to fail
exhibit the highest failure rates and/or cause • Preventive maintenance, i.e. overhauling items or
unacceptable power interruption impact. Without doubt replacing components at fixed intervals
the transformers and circuit breakers are the prime • Corrective maintenance, i.e. fixing things either
candidates for these kinds of monitoring systems. when they are found to be failing or when they have
The second trend within the data acquisition falls into failed
the category of intelligent electronic devices (IED), i.e.
secondary equipment like protection terminals. Besides
their primary functions, they host more and more
additional functionality, which increase their
attractiveness compared with dedicated single function
units. (Figure 3) Many of these additional functions
provide a sound foundation for basic monitoring
systems, cost-efficient and perfectly suited for medium
and distribution voltage level IEDs for protection or
control may comprise:
• Disturbance recorders
• Event recorders
• Statistical value recording (peak current indicators,
number of starts/trips, current at tripping, etc.)
• Power quality analysers
• General purpose programming capabilities that allow
to write and run customer specific applications on the
IEDs.

Automatic printing

Advanced analysis

Summary report GPS

tools

User friendly

visualization

Universal Time

CONCISE / FAST synchronization

Alarm Classes
Distance to Fault

Sequence of Events

# Of trips IED Parameter

Figure 3: Intelligent electronic device for protection

Reliability Centred Maintenance

The new approach is to move from the traditional time-

based maintenance policy to condition -based reliability
• Detective maintenance, i.e. to detect hidden failures by
means of special functional checks and
diagnostics
The type of maintenance policy to select for specific
equipment for transmission and distribution depends on
reliability and on economic and customers’ business
related availability considerations, which take the
consequences of failures into account.

Integrated Transformer Monitoring

Transformer unavailability has a considerable impact on

the operation of electricity generation and installation
networks. Monitoring systems available on the market
already provide solutions for monitoring the windings and
magnetic circuit (e.g. gas dissolved in oil, ultrasound
emission, temperatures). Other methods and systems are
under development. They all aim to prevent major failures
and extend service lifetime of the equipment by triggering
preventive maintenance. Furthermore, the use of
monitoring data must be coupled with the implementation
of diagnostic tools.

SA supports transformer on- line to survey specific

condition related data over a long period of time to enable
the judgement of the transformer condition with respect to
load carrying capability, ageing and development of
failures. These data are retrieved from suitable instrument
transformers and sensors as shown in figure 4.

Bukholz gas accumulation

Ambient temperature
Leakage current
Load Current
Temperature Top

Fans

Currents

Pumps

Humidity:Aquaoil

Temperature Bottom

Dissolved gases: HYDRAN

On-load tap changer position
On-load tap changer temperature differential

Figure 4: Retrieval of transformer condition related raw data

Condition Related Data

The data measured on new transformers are the most

important reference data for the later judgement of the
actual condition. The most common sensors for on-line
condition monitoring are:
• Moister sensor: The presence of moisture in the
transformer oil may be an indication of accelerated ageing
of the insulation or an indication of gasket deterioration
and inward leakage.
 4/8

Figure 5: Access to transformer data via Internet browser

• Top oil temperature sensor: Top oil temperature is
an indicator of the thermal performance of a
transformer.
• Combustible gas sensors: Thermal and electrical
stresses breakdown the dielectric oil into a variety of
gases. The gases are indicative of developing faults in
the transformer and early detection will trigger
corrective action to prevent costly failures. Such sensors
are most sensitive to hydrogen and carbon monoxide.

Transformer Monitoring via Intranet

The World Wide Web as an information source and its

commercial application has created a mass market
where the technology costs are shared by millions of
developers and companies throughout the world. The
costs for applying these mainstream technologies to
other applications like transformer monitoring are
relatively low when compared with proprietary
solutions for networks, protocols, processors, and
software environments.
As an alternative to the integration of on- line
monitoring in SA system a stand-alone solution for
global access for service and support via the corporate
Intranet.
On the server side in the substation condition related
raw data are retrieved from the transformer continuously
in an asynchronous mode and stored in a database
management system (DBMS). The client is installed on
a PC in the maintenance department or on a notebook
PC of the maintenance engineer. A standard browser
e.g. from Netscape serves as human machine interface.
The utility homepage contains a button for transformer
monitoring, which can only be activated by authorised
personnel. If this page is called up, a Java applet is
loaded which establishes the connection with the
hypertex transfer transmission protocol (HTTP) server
on site and the transformer monitoring page is refreshed
with actual data from the server database. (Figure 5)

Vendor Service &

Public

support

Internet

Transformer

Corporate

Intranet Maintenance

Department

HTTP Server

Netscape browser

CGI

Condition related

Gateway

Row Data

Maintenance

DBMS

Server
If a problem is detected, which cannot be judged by the
utility maintenance staff, the vendor’s support and
service centre can be contacted via the Internet

Information Available via Browsers

The transformer-monitoring page on a browser shows

the following actual service and condition related data.
(Figure 6)
• Service data
• Oil temperature
• Current
• Voltage
• Load
• Tap changer position

Figure 6: Web site for transformer monitoring

• Service condition status (ok/not okay)

• Air channels
• Cooling performance
• Oil pump
• Oil temperature
• Overall condition
• Service condition data
• Fan performance in %
• Air entry/exit temperatures
• Oil entry/exit temperatures
• Oil pump switched on/off
• Life expectancy related information
• Load in % of the rated load
• Estimated hot-spot temperature
• Ageing ratio
• Estimated consumed life time in %

Retrofit for Critical Transformers

How critical is the status of an old transformer? Is it

really necessary to shut down its operation to find out?
Periodic, off-line tests play an important role in
evaluating the general condition of transformers. But
more and more utilities wish to have better information
and are turning to a sophisticated process to collect
information while the equipment is still in service. This
continuous monitoring of the transformer aims at
improved reliability, at early stage detection of
problems, and at reduced maintenance cost.
 5/8

But in the majority of cases, there are neither critical line circuit breakers (CB), a transfer bus can be
communication links nor suitable sensors available for provided and the bus coupler CB can temporarily
transformer monitoring from remote, which makes. substitute the faulty line CB (Figure 8). This means,
remote monitoring is not very cost effective. To however, that the bus coupler needs to be equipped not
improve this combination with remote control of the tap only with busbar current differential protection P C but
changer, load dependent control of the cooling system, also with distance protection P L, which can be adapted
and for load adaptive protection enables optimised to the impedances of the various lines.
transformer load control and better utilisation of the
transformer. Internet technology, however, provides The corrective measures, which are automatically
affordable communication facilities e.g. via telephone initiated in case faulty line circuit breakers (CB) to
networks. A combined transformer monitoring, restore the power, would be:
protection, control and communication package is 1. Isolation of the line CB
shown in figure 7. 2. Initiation of all by-pass connections
3. Activation of the correct distance protection

Line Bay 1 Line Bay 2 Bus Coupler Transformer Bay

Intranet
with By-pass with By-pass
Critical transformer

GLASS Server
Transfer Bus CB

CB CB *) CB
Transformer protection
PL PC
Tap changer control

Condition related data Cooling system control P

PL PL T/G
Protection data

*) enabled only if

bus coupler CB is

used as line CB

PL - Line Protection

PC - Bus Coupler Protection

PT/G- Transformer/Generator Protection

CB - Circuit Breaker

Figure 7: Retrofit package for critical power transformer

Distance line protection with transfer bus

INTEGRATED PROTECTION AND CONTROL
With the introduction of numerical relays, with the The power supply from the public network into an
realisation of modern substation control systems, and industrial plant is usually highly sensitive against
with the development of new communications interruptions, as the consequential damages caused may
technologies, direct interaction between protection and be exorbitant. In order to provide redundancy for the
control leads to adaptive protection schemes, to
intelligent load shedding, and to automatic power
restoration procedures. The following examples
illustrate how these new functions can be implemented
in conjunction with SA concepts as corrective measures
to limit the consequences of faults.

Adaptive Protection

The term “adaptive protection” is related to a protection

philosophy which permits and seeks to make
adjustments in various protection functions
automatically in order to make them more attuned to
prevailing power system conditions. This means that the
function of a protection relay, measuring local process
related quantities can be enhanced by additional
information about the network.
 Figure 8: Distance Protection with Transfer Bus

Compensation of mutual impedance

The next critical areas of failures are the redundant

power lines feeding into an industrial plant, which often
run in parallel over long distances. The automatic
switching of the load from one line to the other as a
corrective measure in case of one line being faulty, has
to take into account that mutual impedance exists
between the parallel lines. This impedance can cause
measuring failures, resulting in unnecessary trips
initiated by the associated distance relays during earth
faults. In order to avoid this, the distance protection
needs to be automatically adapted to the topology of the
parallel lines and to the actual service conditions (e.g.,
parallel, disconnected, earthed or unearthed, both lines
connected to different busbars at one side, etc.) . Apart
from this, also the power carrying capability of one of
the lines may have to be increased by corresponding
adaptation of the line protection.

The scheme for the corresponding exchange of

information between the line bay control units and the
line bay protection units as well within the substation
itself as between associated substations is shown in
figure 9. It is crucial that this communication is of very
high quality with regard deterministic and real time
speed behaviour. It is therefor recommended to
 6/8
Actual load

Actual busbar voltage

establish a communication link, which is dedicated for current per feeder

this adaptive protection task. and frequency (f)

Communication between substations

Line distance Line bay Line bay Line distance

protection control control protection

Station 1 Station 2

Line 1
the
Communicationwithinsubstation

Communicationwithinthesubstation
Line 2

Line distance Line bay Line bay Line distance

protection control control protection

Communication between substations

Figure 9: Topology based adaptation of line distance protection

DYNAMIC LOAD SHEDDING

Load shedding is a typical corrective measure to assure

uninterrupted power supply to vital areas of an
industrial plant, if there is an under-frequency condition
due to lack of power.
The conventional load shedding approach is static, as it
initiates tripping of pre-selected circuit breakers when a
certain level of under-frequency is reached, regardless
of the actual load conditions. The reason is that the
actual load behind each individual circuit breaker is not
taken into account.
Microprocessor based load shedding schemes, however,
are in the position of considering the actual loads and to
dynamically select only those feeders to be opened,
which are needed to regain the frequency stability

Dynamic Load Shedding Dynamic selective

feeder tripping

Selection Release according to actual loads

Adaptive selective

shed table decision

Pref

Transmission

X% network

P P1...Pn > df/dt <f

I1..I n U automatic load Busbar

Priorities shedding
 class will not be enabled and the load shedding scheme
is reset as soon as the network frequency has recovered.
communication links from the CT’s and VT's, which If the network frequency has recovered, the integrated
are incorporated in the numerical protection/control network restoration function will be started
devices. automatically.
The LSFB compares the reference power Pref with the f (Hz)
individual feeder load measurements P1...Pn. To each Disturbance

feeder a priority index Pr is assigned for load shedding. f

The LSFB selects from the power inputs P 1....Pn the N

sum of the power which is larger than Pref thus

minimising the difference between the selected and f

reference power. If the pre -determined load shedding Lim 1

criteria (LSC) in terms of under-frequency (< f) or f

frequency change (> df/dt) is fulfilled, a predefined Lim 2

percentage X % of total load Ptot is shed by opening

selected feeders. The selection of the feeders to be t (s)

opened also takes the predefined priority index P r into P (MW)

account.
Load of priority 1 X % of total Load

If the network frequency continues to drop or remains Step 1

stable on an under-frequency level, the shed of the next
load class is initiated, i.e. shedding of a second New load Step 2 Load of priority 2 Y % of total Load

balance
predefined percentage X% of the total load Ptot (Figure
11). Otherwise, if the network frequency starts to t (s)

increase within a definable time delay, the next load

Distribution
network Figure 11: Dynamic Load Shedding

Figure 10: Dynamic load shedding scheme

Benefits
Load shedding functions can be allocated to a protection
or a control device associated with the various bays of In contrast to the conventional way of load shedding,
an SA system. stabilisation of the frequency can often be reached in the
first shedding step. In addition, only the necessary load
The load shedding function block (LSFB) of the
is tripped resulting in a minimum impact for the plant
dynamic load shedding scheme continuously monitors
supply.
the load of each feeder. (Figure 10) It obtains the actual
measured current and voltage values either directly
hardwired from dedicated CT’s and a busbar VT or via
 7/8

• Bay oriented remote control

• Integrated bay oriented numerical protection
Network Restoration • Retrieval of data from numerical protection
Another important feature of the microprocessor based • Communication between RTU and IEDs (Peer-to-
load shedding is selective network restoration, which is peer mode)
comprised of two steps: • Support of fast interaction between IEDs
1. Generation increase: If the load balance is reached • Automated intelligent power restoration
the network frequency will get stable on a lower level. • Retrieval of condition related data
To recover the network frequency, activating the • Maintenance server with direct access
generation reserve will increase power generation. The
load frequency controller leads the network frequency Maintenance NCC/SCADA

back to the nominal value. Centre

2. Network restoration: The load shedding program Local Workplace

has stored the tripped feeders. After the restoration of

the frequency the feeders are reconnected one by one.
Each reconnected feeder causes a minor load unbalance, Control room

which leads to a certain frequency reduction Switchyard

immediately restored by the load frequency controller. Interbay Bus

This frequency supervision avoids network collapse

during the restoration phase.
*) *) *) *)

DISTRIBUTED SA VERSUS RTU X X X X Cabeling

The remote terminal unit (RTU) concept as shown

Figure 13: Distributed integrated SA concept
in figure 12 comprises the following features:
The comparison between the range of functionality
Centralised RTU makes it obvious that the only the distributed integrated
• Communication to network control centre SA concepts offers the benefits for more power by
• Bay oriented remote control supporting the issue for better information, which leads
• Bay oriented numerical protection to higher productivity, as well as the automation aspect
to achieve higher availability.
• Retrieval of data from numerical protection
• Communication between RTU and IEDs WIDE AREA PROTECTION
(Master/slave mode)
• Support of IEC 870-05-103 protocol A new technology has been developed for the detection
NCC/SCADA
of incipient voltage instabilities. Such systems comprise
a number of Phasor Measurement Units (PMU) and a
central evaluation unit. The PMUs are located at
strategic points throughout the power system. (Figure
13) They contain an algorithm that calculates the
Remote Terminal Unit (RTU) stability of the various power line transfer conditions
Serial Master/Slave Bus
based on local measurements of voltage and current
phasors.
Relay room

Switchyard

Re

Cabling
U1 I1

X X X X
I3

U3 Transmission Network

Figure 12: Centralised RTU Concept Im

PMU U2

I2

As an integrated alternative, the distributed SA concept System

as shown in figure 13 comprises the following features: Protection

Centre
Distributed integrated SA Re

U1 I1

• Communication to network control centre I3

U3 Im
 U2

I2

Determination of

relative rotor

PMU

angle differences

PMU PMU

Re Re Re

U I U I
1 1 1 1 U1 I1

I
I3 I3 3

Im Im Im

U U U U
U3 2 U3 2 3 2

I2 I2 I2
 8/8

Response to Incipient Voltage Instability Decision Support Systems

In case of incipient voltage instability During disturbed condition data from disturbance
detection, two types of instructions are issued: recorder, information on network configuration,
protection relay signal data and operational procedures
1. Alerting the system operator by indication of the
all need to be considered when arriving at the
remaining safety margin S and by providing on-line
appropriate decision for corrective action. This can be
guidance to counteract this situation. In addition,
problematic due to data volumes and the limitations to
corresponding information is produced as input for the
the amount of information and real- time performance,
energy management system (EMS).
which can be managed from SCADA and handled by an
2. Automatic control actions are initiated if the operator. The advantages of an automated system
safety margin S reaches a pre-set critical level to located in the substations, which accelerates the
avoid voltage instabilities to occur, e.g. automation process, are obvious. The corresponding
system structure below shows the allocations of
• To compensate the lack of reactive power by
functions and the communication links.
FACTS (Flexible AC Transmission Systems)
• To block the OLTCs (On-lin Tap CONCLUSION
Changer Controller) in order to avoid over
compensation. The SA infrastructure comprising numerical
• To maximise the reactive power output protection and control devices allows a new
from synchronous machines, to compensate maintenance approach for T&D equipment and
the lack of reactive power. provides cost effective solutions for
• To shed load as last defence measure
• On-line monitoring of primary and
The block diagram below shows the secondary substation equipment
interactions between the various applications.
• Diagnostics to determine the need for maintenance

System Restoration

U
EMS
The two objectives to be achieved during
I PMU
any major disturbance are
Phasor 1. To preserve the integrity of the transmission grid
angle by shedding load and/or generation as required
difference
U VIP System Operator
2. To restore supply to all consumers in the
Phasors Phasor Voltage
∆S
shortest possible time
I

With an effective islanding system in place

PMU Evaluation Instability Guidance
considerable advantage can be obtained from
(Display)
well-designed auto-restoration schemes.
Predictor
Integrated numerical control and protection
U
devices in a substation automation system enable
PMU
I ∆S
to design intelligent automated restoration
schemes.

Automated control actions

Automated

Control

Tap

Load

Shedding AGC Changer FACTS

Control ∆S : Safety margin as

proximity to voltage collapse

Load Generation Tap Changer Cosϕ

Adaptation Adaptation Blocking Adaptation

• Reliability-centred maintenance and
condition based asset management
The benefits are reduction of operation and
maintenance costs as well as improved service
quality and overall power system availability
The integration of protection and control is a way to
minimise installation and maintenance complexity by
reducing the number of IEDs in a substation.
Thorough research studies have indicated that the
integration of protection and control functions will
even increase the system reliability. The improvement
arises largely from the reduced amount of hardware,
wiring and interfaces between equipment that are
enabled by integrated solutions.
This conclusion has a very important impact on
substation automation and power system
monitoring concepts as it makes a combined
protection, control and monitoring concept not
only economical but also superior in reliability in
comparison with a centralised RTU concepts and
separate IEDs control and protection.

A further advantage of this approach is that automated

responses on incipient system faults are enabled. Large
area outages caused by voltage instabilities and
frequency declines can be counteracted by means of
wide area protection schemes, adaptive protection for
transmission lines, intelligent load shedding and
automated defence plans power system restoration.