448

Refurbishment Strategies based on

Life Cycle Cost and Technical Constraints

Working Group
B5.08

February 2011
 Refurbishment Strategies based on
Life Cycle Cost and Technical Constraints

Working Group
B5.08

Members

Peter Leushuis, Convener (NL), Jokin Galletero Lopez (ES), Daniel Garcia Garcia (ES)
Jean-Michel Grellier (FR), Frank Koers (NL), Thomas Küng (CH), Mika Loukkalahti (FI)
Jerzy Pillar (PO), Didier Wiot (BE), John Wright (GB)

Copyright © 2011

“Ownership of a CIGRE publication, whether in paper form or on electronic support only infers right
of use for personal purposes. Are prohibited, except if explicitly agreed by CIGRE, total or partial
reproduction of the publication for use other than personal and transfer to a third party; hence circu-
lation on any intranet or other company network is forbidden”.

Disclaimer notice

“CIGRE gives no warranty or assurance about the contents of this publication, nor does it accept
any responsibility, as to the accuracy or exhaustiveness of the information. All implied warranties
and conditions are excluded to the maximum extent permitted by law”.

ISBN: 978-2-85873-137-4
Table of Content

Introduction ....................................................................................................................... 4

1 Definitions ........................................................................................................ 5

2 Secondary System Evolution, Levels and Position ........................................... 7

2.1 Technology Evolution ....................................................................................... 7
2.1.1 Classical to Static............................................................................................. 7
2.1.2 Static to First Generation Digital....................................................................... 7
2.1.3 First Generation Numerical to Numerical Today ............................................... 8
2.1.4 Hybrid System (Conventional / Numerical System) .......................................... 8
2.2 Secondary system as a part of the overall system ........................................... 9
2.2.1 Overall system ................................................................................................. 9
2.2.2 Description of secondary systems .................................................................... 9
2.2.3 Process level: ................................................................................................ 10
2.2.4 Bay level ........................................................................................................ 10
2.2.5 Station level: .................................................................................................. 10
2.2.6 Network / Grid level: ....................................................................................... 11
2.2.7 Purpose of secondary system ........................................................................ 11
2.3 Power system market..................................................................................... 11
2.3.1 Economical reasons for substation automation .............................................. 12
2.3.2 Changes in the power system market ............................................................ 12
2.3.3 New Nature of Power System Market Requirement ....................................... 13

3 Driving forces (Triggers) for change ............................................................... 14

3.1 Extension of the station .................................................................................. 14
3.2 Equipment obsolescence ............................................................................... 14
3.3 Reliability and availability ............................................................................... 15
3.4 Lack of competence and documentation ........................................................ 16
3.5 Maintenance savings ..................................................................................... 17
3.6 Functional requirements and added value...................................................... 17
3.7 Data requirement ........................................................................................... 18
3.7.1 Functional demands and performance enhancement ..................................... 19

4 Refurbishment Strategy ................................................................................. 21

4.1 Risk management introduction ....................................................................... 21
4.1.1 Identifying all possible risk ............................................................................. 22
4.1.2 Catalogue the risk in different domains .......................................................... 22
4.1.3 Assessment of the identified risk .................................................................... 23
4.1.4 Identify the objectives of the stakeholder........................................................ 24

1/80
4.1.5 Analysis of risks assessment and objectives of stakeholders ......................... 24
4.2 Financial models ............................................................................................ 24
4.2.1 TCO ............................................................................................................... 25
4.2.2 Net Present Value .......................................................................................... 26
4.2.3 Internal Rate of Return ................................................................................... 27
4.2.4 Return on Investment ..................................................................................... 28
4.2.5 Payback Period .............................................................................................. 29
4.2.6 Life cycle cost ................................................................................................ 29
4.3 Summary ....................................................................................................... 31
4.4 Examples ....................................................................................................... 32
4.4.1 Introduction .................................................................................................... 32
4.4.2 Example 1: high risk of existing system ......................................................... 33
4.4.3 Example 2: high benefits new system ............................................................ 34
4.4.4 Overall methodology ...................................................................................... 35

5 Options and Strategies for Change ................................................................ 37

5.1 General .......................................................................................................... 37
5.2 Migration Options ........................................................................................... 37

6 Utility feedback – approach & practices ......................................................... 40

6.1 Summary on Questionnaire............................................................................ 40
6.2 Practical Experiences in Some Countries....................................................... 42
6.2.1 United Kingdom approach .............................................................................. 42
6.2.2 Finland approach ........................................................................................... 43
6.2.3 Poland approach ............................................................................................ 44
6.2.4 France approach ............................................................................................ 50
6.2.5 Belgian experience ........................................................................................ 54
6.3 Conclusions ................................................................................................... 59

7 Trends ........................................................................................................... 61
7.1 System trends ................................................................................................ 61
7.1.1 Increasing network load ................................................................................. 61
7.1.2 Unpredicted, changing load flows in the network............................................ 62
7.1.3 Reducing investment and operational cost (total cost of ownership) .............. 62
7.1.4 Improving reliability ........................................................................................ 63
7.1.5 Reducing technical skills within the company (outsourcing of non core activity)
...................................................................................................................... 63
7.1.6 More information from the status of the network to management, load forecast,
systems, etc. .................................................................................................. 63
7.1.7 Scale and amount of blackouts is increasing. ................................................. 63
7.1.8 Flexible networks ........................................................................................... 63
7.2 Self check contribution to reduced maintenance ............................................ 64

2/80
7.2.1 Self supervision definition .............................................................................. 64
7.2.2 Scope of Self supervision ............................................................................... 64
7.2.3 Coverage ....................................................................................................... 65
7.2.4 False alarms .................................................................................................. 66
7.2.5 Self action (response from the device) ........................................................... 66
7.2.6 Prevent malfunction ....................................................................................... 67
7.2.7 Analyse signalling what information do we need (for efficient maintenance
policy) ............................................................................................................ 68
7.3 Remote maintenance ..................................................................................... 69
7.3.1 Remote setting ............................................................................................... 69
7.3.2 Measures for security ..................................................................................... 71

8 Conclusions ................................................................................................... 72

9 References .................................................................................................... 74

10 Appendix ........................................................................................................ 75
10.1 Maintenance strategy. Example 1. ................................................................. 75
10.2 Maintenance strategy. Example 2. ................................................................. 75
10.2.1 Example of maintenance measures ............................................................... 76
10.2.2 Functional requirements and added value...................................................... 77
10.3 Examples of self-check functions ................................................................... 78
10.3.1 Hardware ....................................................................................................... 78
10.3.2 Software......................................................................................................... 79
10.3.3 IEDs response and recommended actions ..................................................... 80

3/80
INTRODUCTION

Many utilities around the world are confronted with the fact that a number of their
secondary systems are approaching the end of their technical life. Such systems include
a mix of technology from “old” electromechanical and analogue systems to first generation
digital systems. Utilities are faced with the decision to either extend the working life of the
existing systems, by say upgrading, refurbishment etc or totally replacing them. Drivers for
the refurbishment of secondary system(s) include no spare parts, lack of knowledge,
missing documentation and the introduction of new functionalities, such as self-check.

The goal of this report is to propose a practical guideline to assist asset managers when
making decisions related to refurbishment strategies. The triggers for such decisions are
often operational risk or technical constraints and will be typically followed by a Life Cycle
Cost (LCC) calculation to determine the financial impact.

Practical examples of refurbishment strategies will be given in chapter 6.2.

The working group distributed a questionnaire to a number of utilities, the results of which
are detailed in the appendix.

4/80
1 DEFINITIONS

Due to the varying interpretations of the terminology used in this field of Engineering, the
working group deemed it necessary to define the following terms for the purpose of the
report.

System Level (Bay or Substation)

Replacement: is the process of replacing the majority of the existing infrastructure with
the intention of improving the remaining lifetime and the functionality.
Example: replacement of a complete protection cubicle retaining the
interfaces to the primary system, i.e. CT’s, VT’s.

Refurbishment: is the process of major maintenance or minor repair to the existing

infrastructure with the intention of improving the remaining lifetime at a
minimum cost.
Example: Replacement of a number of protective relays within one or
more bay while retaining the existing cubicles.

Retrofit: refers to the addition of new technology or features to older systems

with the intention of improving the functional level at a minimum cost.
Example: replacing old electromechanical relays by new digital relays
(existing cubicles are fitted with new digital relays) often using the same
housing.

Device Level

Replacement: replacing of a device by one of a similar type or functional equivalent.

Repair: restoring of the device functionality after failure

Upgrade: improving of the device functionality or performance (firmware, software

or hardware)

The following table shows the approximate / anticipative effects of each type of change
that is described in this report.

5/80
Table 1: Change options classified by level

Remaining Lifetime Functional Level

Definitions Unchanged Improved Unchanged Improved Risk Cost
Replacement X X High High
Substation
Refurbishment X (X) (X) Medium Low
Level
Retrofit (X) (X) X Low Low
Replacement X (X) (X) Low High
Device
Repair X X Low Medium
Level
Upgrade (X) (X) (X) (X) Low High

X = driving force

6/80
2 SECONDARY SYSTEM EVOLUTION, LEVELS AND POSITION
This chapter describes briefly the technical evolution of secondary systems, the different
levels in a substations and the position of the secondary systems in the network. In order
to assess a chosen strategy, decision makers need to understand these items.

2.1 Technology Evolution

Technology evolution, from electromechanical to numerical systems, has brought

considerable economical and technical benefits. The price of such numerical devices is
comparable or less than electromechanical devices. The numerical systems offer more
functionality and information about system status. With modern systems, maintenance
intervals can be extended and onsite commissioning can be reduced. The numerical
system can be extensively tested at the factory, thereby reducing testing on site. In the
next chapter the technical evolution of secondary systems are briefly described.

2.1.1 Classical to Static

Traditional protection schemes consisted of many electromechanical relays all

interconnected by hardwiring. The relays were typically housed in multiple protection
cubicles, taking up a considerable amount of space in the substations. Classical systems
required a considerable amount of installation and commissioning effort. The schemes
were inflexible, changes required additional relays and wiring modifications. Due to
technology development, electromechanical relays were replaced by static relays. The
static relays allowed a reduction in CT requirements and reduced in size, resulting in
space saving. However they require a DC supply. The static relays were still one box, one
function, hence a considerable amount of hardwiring was still required.

2.1.2 Static to First Generation Digital

With the introduction of multifunctional numerical relays, protection refurbishment started

to increase since a number of benefits were obvious:
− Reduced panel space due to multifunctional relays
− Reduced hardwiring; replaced by software coding
− Self monitoring; CB Status
− System monitoring
− Fault recording
− Disturbance recording
− Reduced on site testing since scheme can be simulated in the factory
− Ease of interface to SCADA & DCS
− Possibility to simplify or automate test on site

7/80
It is important that the skills of the engineers are in line with the technology. Engineers
have to be trained to program and maintain such devices. They can be complicated, and
the management of software files and schemes can be time consuming and difficult.

2.1.3 First Generation Numerical to Numerical Today

Some of the most important features of a protection device are reliability, ease of setting
and testing. Numerical systems offer a large amount of protection functions in the same
box, such as; self supervision, fault recording and programmability Drawbacks of this new
technology are; complexity and version handling of software. Hardware depends on the
newest PC technology and software bugs (which fortunately are minor).
Issues related to the first generations systems are;
− Possibility and cost of upgrading or renewal of first generation systems (not at the
end of their lifetime),
− Ageing of digital systems (especially PC-based ones),
− Update policy for digital systems (step-by-step or only when needed - if this is
possible), version handling requirements for tools, PC's, operating systems and
vendor software. In early digital products, the upgrade ability was limited.
Compatibility between some relay setting tools and early digital products have
been problematic. This requires a lot of attention from the user.

2.1.4 Hybrid System (Conventional / Numerical System)

In many substations around the world protection systems are hybrid, i.e. a mixture of
different technologies. This mixture of technologies results in the following:
− Engineers require training in different relays and different technologies
− Mixtures of auxiliary voltage levels
− Different environmental operating withstands
− Different ratings
− Different CT / VT requirements
− Mixture of traditional scheme drawings and internal software scheme logic.

Due to the mix in systems; some points need to be addressed, such as;
− Trip relays required or trip breaker direct?
− External or internal trip circuit supervision?
− Can the numerical relay exactly simulate the old scheme?
− Is the old scheme still valid or can it be rationalized?

8/80
2.2 Secondary system as a part of the overall system

The secondary system is the heart of overall power system. Auxiliary power, protection,
measuring, control and monitoring systems are key elements to the successful operation
of the total system. The protection and control system is maybe the most demanding and
difficult part of it, as it includes the user / operator interface to the whole installed system
for control and support of operation philosophy.

2.2.1 Overall system

The overall system covers all equipment to generate, transmit, distribute, control and
ensure electrical power. This is divided into primary and secondary system.
Separation of primary and secondary system:

Network Ctrl Station Ctrl Bay Control Drive

Drive
RTU Prot. Main 1 Drive
Gateway Prot. Main 2

Station Prot. Disturb. Rec. Drive

Monitoring
Drive
Metering
Drive

Network Level Station Level Bay Level Process Level

Secondary System Primary System

Figure 1 Overview overall system

The primary system covers the high voltage part of the system on which the electrical
power of the network is flowing. It covers overhead lines, transformers, circuit breakers,
disconnectors, instrument transformers (CT’s and VT’s), surge arrestors, reactors, etc.
The electrical part for some of the primary devices, like the drives of circuit breakers,
transformer / tap changer control boards and disconnectors are considered part of the
secondary system

2.2.2 Description of secondary systems

The secondary system covers all electrical parts, which support a utility / grid operator to
control and ensure the flow of electrical power. To understand and validate its value the

9/80
secondary system itself is subdivided in 4 hierarchical levels:
− Process level
− Bay level
− Station level
− Network / grid level

Depending on the technology used, functionality from more than one level can be
allocated in the IED (Integrated Electronic Device) e.g. Station controller controls host of
several bay control functions.

2.2.3 Process level:

All secondary equipment to control, monitor and protect single primary devices , such as;
motors, contactors relays, energy limiting switches of circuit breakers and disconnectors,
transformer / tap changer control box directly mounted at the transformer and secondary
wiring between primary equipment and bay cabinets

2.2.4 Bay level

All secondary equipment to control, monitor and protect one bay / feeder / unit.
Bay controller including interlocking and synchronization equipment respective
synchrocheck equipment
− Bay protection
− Revenue meters
− Monitoring (online: display of actual operation value, e.g. volt, amp, freq. and
power meters)
− Disturbance recording per bay

2.2.5 Station level:

All secondary equipment to control, protect and monitor a substation:

− Station controller (sequencer for control automatic functions)
− Station control system (conventional: electromechanical or numerical:
microprocessor / PC based)
− Gateway
− RTU’s
− Station protection (bus bar protection, load shedding functionality, high speed bus
bar transfer functions)

10/80
 − SOE (sequence of event list)
− Disturbance evaluation

2.2.6 Network / Grid level:

Network management system:

− Network control system (main user interface for the grid operator)
− Network information system
− Network maintenance system
− Telecommunication between network and station level.

2.2.7 Purpose of secondary system

The secondary system can be seen as an interface between the primary system and the
network management system. Acquisition cost of the protection and control assets is
generally less than the price of the primary assets. Still secondary systems need more
engineering work than primary system: parameterization, setting and commissioning is
time consuming work. On the other hand the failures especially in protection systems
(false trips or unselective trips) are very significant and unwanted. Also every primary fault
has to be detected by the protection and control systems. With a good secondary system
you can detect and isolate the primary faults fast and selectively and obtain the right
information about them. However, you cannot prevent primary faults with protection and
control systems.
In principle, the secondary system could not totally prevent the risk of damages in the
primary system. But it helps to minimize the direct damages of primary equipment failures
by interrupting power as well as reducing the consequential damages out of a grid failure
e.g. in case of a busbar fault.

2.3 Power system market

The power system industry is working in an environment that requires optimal

management of the power system network at all system levels. At transmission level
companies must ensure a stable and optimal operation of the grid, whereas in other
voltage levels there is growing competition in the market place. The privatization of the
power system industry creates the opening for a new electricity market differing in all
aspects from the traditional old market. A market where the consumers become
customers is due to that new energy supply and traders are appearing in the market. In
fact, in the very near future, power system industries all- over the world will see more and
more power producer, retailers and network companies.

11/80
Therefore, the need to automate the existing substations shall be evaluated by the utilities
in order to meet the expected challenges of the future market and the reliability of the
existing equipment.
Each utility shall first of all, prepare itself and its network for the challenges ahead. In
order to do this, the utility must acquire full knowledge of its automation needs and
benefits. The utilities in their effort to automate the existing substation shall focus on two
aspects that shall influence the optimum control of its power system management
business. These two aspects are economical and technical.

2.3.1 Economical reasons for substation automation

The economical reason plays a major part in justifying existing substation automation
(SA). The information about the power system gives the utility the strength to be more
successful and competitive in a free market where competition between utilities and the
deregulation of the power system industry is being introduced. In this type of environment
the information becomes a very strategic requirement in the power system industry market
where a fast decision is required. This cannot be obtained from an existing conventional
type substation. The changes, which are occurring presently and expected in near future
in the power systems industry can be listed as follows.

2.3.2 Changes in the power system market

Major changes have taken place in the power system market and more are expected in
the future. The traditional market where Nation/Area-wide power control centres play the
role of control and marketing the energy, since there were no other suppliers of electricity
to the customer, is disappearing gradually and this trend will continue at a faster rate.
Energy service Companies are replacing power system companies and new retailers of
energy are being introduced in the market. Also, the privatization/deregulation of the
nations electrical networks find interest for even non-national Companies in the market.
This gives the chance to have a new power producer and retailers. Such type of market
shall be an open type where the consumer is becoming a customer and he can choose
his supply contractor, increase the competition in power system efficiency while
maintaining system security and reliability. A market having a variable electricity price at
present and in the future depending on the market competitions as what can be said
Electricity at the Market price.
Activities of transmission and distribution utilities have to separate on open market
business (energy market) and regulated business (transmission and distribution). On the
energy business there are new functions required for running energy market, while on the
regulated field, there are no major changes regarding required functions. There is only the
need to provide all necessary data for supporting energy market activities. It is not clear,

12/80
who should cover the costs, if this functionality would exceed equipment installed for utility
use.
This type of environment requires that the existing substation shall be upgraded to provide
the necessary urgent information. This is also in addition that the market must meet the
loads needs.
Other additional functions in SA have to be economically proved by other factors, than
new energy market situation.

2.3.3 New Nature of Power System Market Requirement

As mentioned previously, the Electricity Consumers in the new market become the
customers. These customers will have the choice to get power supply from different
suppliers that are geographically spread.
Therefore, new types of power supply agreements shall be introduced in the market.
These power agreements shall handle the power supply at different interval of different
prices and different suppliers. This is what is called a free market price and place.
Suppliers provide daily information of the power transfer capabilities and retailers receive
consumption information. This requires huge, accurate, fast information of data for billing
the Customers instantaneously. Moreover, the customers also know their daily operational
cost in order to properly plan their production to minimize cost and increase their profit. In
this case, information is becoming a must and existing substations should look to
automation if they are going to a play part in such a market.

13/80
3 DRIVING FORCES (TRIGGERS) FOR CHANGE

Reasons for refurbishment, retrofit, replacement and/or upgrade can be various. Driving
forces can be extension of the primary configuration, equipment obsolescence, failure
risks level, lack of knowledge or maintenance cost. In the next chapters those driving
forces are explained.

3.1 Extension of the station

During its lifetime the system is also likely to change, for example, increased fault levels,
change in load or system expansion. These will all impact on the protection system
design and settings. This action is commonly ignored and has lead to many unnecessary
protection scheme operations and under protected primary plant.
Also, an extension in a substation may be a sufficient driver for replacement of existing
protection relays, even though they would have some remaining calculated life. If the
extension is of significant measure, such a total protection system replacement becomes
even more obvious. This gives a benefit on spares handling, ease of use, documentation
and periodical testing, as all relays at one substation are of same types and versions.
Extension of a substation can also bring along a need to reduce size of new equipment
due to space constraints and may give a final push to replace also the old devices. In
some cases there is a need to reduce energy consumption with modern IED’s due to
batteries dimension requirements.

3.2 Equipment obsolescence

Lifecycle management ensures continual operation without substantial interruption due to

hardware failures, obsolescence or incorrect application. The value of the protection
equipment asset depends on the way its lifetime is managed. Without any particular focus
on this process, the peak value of the asset will be at the initial installation stage, after
which the asset will start to decrease. The more lifecycle support provided the slower the
depreciation of the asset.
For a given technology the cost of spare parts maintenance generally increases with time,
its explanation is that new technologies get a better cost/functionality ratio. Fabrication of
old technology devices is therefore limited to spare parts: the amount of production
decreases and the relative cost of each piece rise.
Also, these types of oldest relays start to become obsolescent and their maintenance may
cost too much. Furthermore, the risk of having a too slow operation of a main protection
leading to non-selective other trips in the HV networks may give a reason to replace such
relays.

14/80
The below table illustrates the key milestones of a products life.

Table 2: example of Lifecycle identification *)

Status Characteristic Approximate

Time
Active The product or system is in manufacture and subject 8 years
to continual development (hardware / software
changes)
Legacy The product or system can still be provided as new, 10 years
but no new features are developed.
Pending The product or part of the system is declared 2 years
Obsolescence ‘Pending Obsolescence’ and customers are notified
of this event. Key actions are identified.
Obsolete The product cannot be obtained as new. A reduced 5 years
set of services remains available (for example repair
but not modify).
Extinct No services are available, except commitments -
related to a specific maintenance contract.
*)
based on manufacturers data

3.3 Reliability and availability

It is clear that over time the reliability of the protection system goes down bit by bit. At the
final stages of their expected lifetime the reliability of protection devices can reduce
significantly. Modern relays have better availability and reliability due to self-checking
function and this reliability improvement need can be impact for changes in the secondary
side of the substation.

The consequences of protection relay failure have the potential to be catastrophic. The
following list details a few:
− Disruption in supplies due to incorrect design and settings or due to relay
malfunction
− Damage to equipment due to excessive fault current resulting from long operating
times.
− Incorrect operation of plant i.e. motors stalling due to low voltages.
− Fire and explosion due to excessive fault current resulting from long operating
times.
− Loss of life or serious injury.

15/80
A system fault may be a rare occurrence, but unless the protection is maintained and
properly serviced then it may not operate correctly. With correct maintenance and
application, should a fault occur, the damage and disruption will be minimised. Therefore
to maintain optimal performance of the protection equipment, it is very important to
manage the life cycle of the protection scheme.
Cost reduction through improved efficiency is the main factor for renovation /
refurbishment of an electrical substation. This is achieved through better network
operation and reduction in the cost of substation maintenance. In general, for a given
technology the cost of spare parts maintenance generally increases with time, the reason
is that new technologies with a better cost/functionality ratio progressively replace the old
ones.
Another important point to comment on are self-checking capabilities; these can
significantly reduce the necessity of periodic maintenance, if such information can be
retrieved remotely. Although self-checking cannot prevent or detect all kind of miss-
operations, self-checking reduces the complexity of some maintenance because some
points are continuously checked by this capability.
The settings in electronic relays may start to change slowly (zone reach in distance relays,
current settings, time settings). This increases the risk of having a slow operation of a
main protection, or no operation at all, leading to non-selective trips in the HV networks.
Thus, this may give a reason to replace such relays.
Due to fast diagnostic of the problems in the system and more information about the
power system the modernization of the secondary devices increases the reliability of the
substations and shortens the period of diagnosis. This resulted in faster restoration of the
substations.

3.4 Lack of competence and documentation

For electromechanical relays one driver for replacement is that the utility may have lack of
documentation, lack of competent and skilled relay engineers and technicians as well as
lack of suitable tools for testing and repairing.
One problem with digital relays is the version handling with quite limited functionalities in
their first versions they are not easy to upgrade into newer versions, at least not without
sending the relay back to the factory to do it. However, this is seldom a driver for
replacement.
Utilities at present are facing difficulties in documenting all changes and upgrades which
are done to the network. In other words, there is no “as built” which reflects the actual site
conditions specifically the secondary equipment. There is a considerable amount of time
wasted in verifying the existing installation before starting any implementation upgrade or
modification to existing installations.
The new IED’s (compatible with IEC 61850) provide self-describing capabilities which can

16/80
be used to create system documentation as a part of the current system implementation.
However in certain cases utilities decide to specify / approve one firmware version for the
product which they purchase. With this approach they reduce problems that could be
derived from having different versions installed. This approach may be recommended for
fixed application without need for extra functionalities.

3.5 Maintenance savings

An examination of the capabilities and functionalities available in modern protection and

control devices may influence the decision to switch earlier than expected to new
schemes to minimise maintenance costs. Self-checking function gives the possibility to
reduce the amount of periodical maintenance.
Maintenance strategies strongly differ from utility to utility. Periodical protection test
interval times vary usually from 1 year to 6 years, old relays need more maintenance than
the new ones. Transmission network protection relays must be secured as good as
possible, so the test intervals at transmission level are usually shorter than in the lower
voltage levels.
Proper and extensive periodical relay testing consists of testing of the total protection
system including relay, measuring circuits, tripping circuits and signalling circuits.
Numerical systems are equipped with measuring and tripping circuit fault detection
however shelf-check functionality does not cover the complete system. Even with self-
check functionality testing is still necessary. The need for complete testing of all settings
and zones is not essential due to digital and numerical technology.
During a test there is always a risk that a human failure is introduced. In some cases test
has to be postponed due to high load of connections in the network.

3.6 Functional requirements and added value

Early protection schemes may not be considered sufficient anymore and thus the
increased performance and functionality capabilities can be a driver for change.
Functional integration capabilities make it possible to group all the feeder functions into a
single device thus replacing a collection of elementary elements. This reduces the cost of
the spare parts and copes with the new functional needs. Other drivers for change may be
equipment standardisation
The new business needs, which require more information, will direct the utilities to
upgrade the existing substations. Therefore information is needed about the industrial as
well as other types of customers, i.e. computerized load forecast, complicated metering
system bulk trading and energy management. The accuracy and the reliability of the data
depend on the utilities/traders. Therefore the data availability gives the utility the chance to

17/80
be strong in a very competitive field. In the next subchapters the major technical issues
that require the upgrading of the existing conventional substation are described.

Examples for need of new functions

In Helsinki Energy (FIN) three most wanted new functions of modern IED’s are: self
checking function, disturbance and event recording and free programmable integrated
functionality. These functions are often also triggers for secondary changes. Self-
checking and disturbance recording functions give the possibility to improve reliability and
availability of protection system. The possibility to integrate many functions into the same
IED gives financial savings, especially at MV level, where separate over-current relay,
directional earth fault relay and auto-re-close relay were used before. At HV level modern
relays provide options to combine synchro-check and backup over-current relay into
differential and distance IED’s.

3.7 Data requirement

Information (or data) plays a very essential part in optimal management of the power
system. Therefore, more and more data is needed continuously to the master control
stations. Data, such as alarms, breaker status, state monitoring, real time acquisition of
measurements (voltage, current, active and reactive power, etc.), energy management
programs availability and energy metering. This huge data requirement, it’s availability
and accuracy is a must in the modern power system industry. Therefore, the power
control centres in the near future will become Information Technology centres. This
requires the existing substation to be upgraded to automated substation to be able to
cope with required information.
Not only the network control centre but other parties are interested in more data.
Protection engineers need to know if their protection has worked as planned, primary
maintenance engineers need preventive monitoring information from the process and
network planning need information about the load. There is a need for information to
coordinate between engineering, planning, maintenance and operation. Separate
maintenance channels connected to the communication bus of the substation automation
system or connected straight to the IED’s could be a solution for separate needs for data.

18/80
3.7.1 Functional demands and performance enhancement

The new modern system offers the addition of new functions to the existing modern
equipment, unlike the conventional existing system which might require considerable
changes in the secondary equipment to add additional functions. The modern control
system provides the chance to obtain a function from different hardware units into the
master station software.
Next the new functionality and performance of modern IED’s are listed, they are divided
into 5 subgroups: Protection functions, Control and automation functions, Fault
information, Communication and Monitoring & supervision.

Protection functions

Sensitivity: Better sensitivity and wider range of settings are possible. Choice of
characteristics is possible for better protection performance.
Ease of operation: Multiple groups of settings are possible which could be remotely
selected or automatically selected. This feature could lead to ease of operations since
manual relay setting changes are avoided for transfer bus operation etc.
Accuracy: Almost no drift in pick up values due to digital techniques. Analogue
measurements (currents, voltages, PF, etc.) with good accuracy can be measured.
Multiple functionalities (integration): Many functions are built into the relay: auto
reclosing, synchrocheck, trip circuit supervision, external trip signals and option to have
multiple protection functions in the same box will help to reduce the number of devices
and simplify panel wiring leading to higher reliability.

Control & Automation functions

Improved local operation: State-of-the-art HMI application provides higher security of

operation, e.g. select and operate functionality
Improved interlocking functionality: Possibility to implement interlocking logics using
data available in the control system
Access to more data: Allows implementation of new or improved automation functions
Power system monitoring: Low frequency, low voltage, load shedding, blackout
functions may be described with formal language of IEC 61131-3 and easily integrated in
any IED
Time synchronization: By using serial or Ethernet communication bus it is possible to
synchronise all IED’s in the substation

19/80
Fault information

Configurable alarms and trip: Alarm and trip/re-close information are programmable and
can be assigned to particular outputs like contacts or LED’s. Group alarms can be
customized as per utility practice/ standards.
Fault targets/ indications: Extensive fault data is available on local display
Fault disturbance record: Built in DFR function eliminating separate fault recorders.
Separate disturbance recorders are not necessary for recording of current and voltage
waveforms. The disturbance data files can be used in simulation of the power system
incidence for detailed analysis by using relay test kits.
Power system fault location: Distance to fault location is built in, helps in faster
restoration of power supply.

Communication

Communication interface: Integration with Substation Automation through serial or

Ethernet communication bus will enable better access to IED -data/information by multiple
actors. Fast communication enables chance to get more valuable information from the
process.
Maintenance channel: Separate maintenance channel or maintenance channel
integrated into Ethernet bus gives opportunity to collect (straight from IED’s) protection
and disturbance information for protection engineers and condition monitoring data of
primary devices for maintenance engineers.

Monitoring & supervision

Supervision: Modern supervision functions provide better availability like IED internal
supervision, trip circuit supervision, measuring circuit supervision, auxiliary voltage
supervision, etc.
Counters and statistics: Modern IED’s can make statistics concerning primary devices
like counting number of primary switching device operations, I2t counters, interruption
statistics, etc.
Trends: Measuring capabilities of new IED’s could give more information of power quality
(e.g. voltage levels, harmonics, sags and swells, flicker), measured temperature trends,
measured gas pressure trends, etc.

20/80
4 REFURBISHMENT STRATEGY

As mentioned in the previous chapters there are a number of different reasons for
replacing an existing secondary protection and control system, namely;
− the risk associated with the existing system is so high that additional steps are
required to improve it.
− the additional benefits of a new system are so high that the overall performance of
the system improves.
− the costs associated with a new system are lower than the system that is in
operation.

System A: Existing system System B: New system

What to do?
Investment cost (CAPEX) Investment cost (CAPEX)
What to do?
Operational cost (OPEX) Operational cost (OPEX)
Risk (high) Risk (low)
Benefit (low) Benefit (high)

Figure 1: what to do system A or system B?

In the next chapter some of the basics related to risk management and financial models
are explained.

4.1 Risk management introduction

Risk management is a methodology whereby in a systematic approach all risk aspects are
analysed. With the results of the risk study a strategy can be developed. This strategy can
be accepting the risk or reducing / controlling the effect of the risk. Risk management is
not only traditional focussed on physical or legal items e.g. disasters, accidents, and
environmental aspects such as fire, earthquake etc. but also on network risk, such as
blackouts. There is also financial risk management that focuses on the financial risk of a
project.
The objective of risk management is to reduce different risks related to a pre-selected
domain to the level accepted by the company or society.

In a risk management process different steps shall be followed, namely:

1. identifying all possible risks
2. catalogue the risk in different domains
3. assessment of the identified risk
4. identify the objectives of the stakeholders
5. analysing results assessment and objectives stakeholders
6. risk treatment

21/80
4.1.1 Identifying all possible risk
This first step is to identify all possible risk. A complete risk assessment and identification
of all possible risks can be very time consuming. It is advisable to start identifying only
those items whereby the risk is very high or the probability of the occurrence of this risk is
very high. If the result of the risk analysis is unsatisfied, then a deeper analysis of all
possible risks might be needed.

4.1.2 Catalogue the risk in different domains

The second step of a risk study is to catalogue the risk in different domains. The domains
are not fixed. For example the domains for a protection and control system can be
SAFETY, ENVIRONMENTAL, IMAGE, REGULATOR and POWER QUALITY. The reason
to catalogue the risk is that in different domains the risk level can be different. It is
necessary to calculate the risk in each domain since the risk level within the various
domains can vary.

The table below gives a brief example;

Table 3: example identification and catalogue risk

Environmental

Power Quality
Regulator
Safety

Image

Description of the risk …

Lacking of knowledge … X X X
Lack of spare parts … X X
Reduced reliability … X X X
Change unwanted power
Reduced availability X X X X X
outage
Missing manuals Long repair time X X X X
…. …

22/80
4.1.3 Assessment of the identified risk

Once the risks have been identified, the next step is to assess the probability of
occurrence and the potential impact.

The standard definition of risk is change multiplied with effect:

Equation 1: risk
Ri = Li p (Li ) )

Rtotal = ∑ Li p (Li ) )
i

Where:
Li potential lost
p probability that the loss will occur

The assessment of the risk is an important step in any risk management study. This step
might start a lot of discussions. The figures needed to assess the probability and impact in
most cases are not based on large databases. In most cases the figures should be more
pragmatically defined. The standard equation of risk is change multiplied by the effect.
This means in theory change and effect have the same priority. Sometimes this method of
calculations does not give the right feeling of calculating a risk. This might be the case at
quantifying risk related to safety. For these items it might be wise to use only the
probability in the assessment.
In many cases it is very difficult to find hard figures. In order to establish a common
understanding of the risk without detailed investigation it might be helpful to reduce the
amount of values that can put in the table. For example:

− Potential lost
o Very low: less than 10 k€
o Low: between 10 k€ and 50 k€
o Moderate: between 50 k€ and 250 k€
o High: between 250 k€ and 1000 k€
o Very high: more than 1000 k€
o
− Risk
o Very low: less than once every 100 year
o Low: between once every 100 year and 25 year
o Moderate: between once every 25 year and 5 year
o High: between once every 5 year and year
o Very high more than once every year

23/80
4.1.4 Identify the objectives of the stakeholder

The next step in a risk management study is to identify the objectives of the different
stakeholders. The stakeholders can be; regulator, the company board, shareholders,
society.
The objectives of the stakeholders shall be combined to one overall objective. The
objectives can be different in the different domains.

4.1.5 Analysis of risks assessment and objectives of stakeholders

The last step in the risk study is to compare the results of the risk assessment with the
identified risk objectives. After the risk is compared, the system that is studied should be
classified in one of the three categories, namely:
The risk level is lower than accepted risk, no further measures are needed.
The risk level is higher than accepted, but with some additional measures the risk can be
controlled (lower than maximal accepted by the stakeholders).
The risk level is too high and the system should be replaced.

4.2 Financial models

To compare different technical systems with each other it is not longer enough to focus on
technical aspects only. The economical aspects related to a project are increasingly
important.
Financial experts have developed different financial models. Some of these financial
models are better in helping the technicians and/or financial experts in the decision
making process. By applying one of the financial models it is possible to compare different
projects or technical solutions with each other. The intention of this report is not to give
detailed descriptions of the financial models, but to support the engineer. Which model is
the most suitable depends highly upon the company policy. This report describes at a high
level different financial models. If the reader would like to investigate the models more
deeply, background information can be easily founded.

The most used financial models are:

o Total Cost of Ownership (TCO)
o Net Present Value (NPV)
o Internal Rate of Return (IRR)
o Return on Investment (ROI)
o Payback Period (Payback)
o Life Cycle Cost (LCC)

24/80
In the next chapters different financial models are explained. As mentioned before this
explanation is just a brief introduction of the model.

4.2.1 TCO

Total Cost of Ownership is a methodology whereby different projects with the same
benefits can be compared with each other. This methodology can be defined as a
systematic quantification of all cost related to a project during the lifetime. For an accurate
comparison of different projects it is important that the calculation period is long enough.
At least one replacement of a system shall be within the calculation period. It is Important
to know that the benefits are not taken into account in the equation.

Equation 2: TCO
t =n

∑C 0 + Ct
TCO = t =1
n
Where:
t the time of the cash flow
n the total time of the project
Ct the net cash flow (the amount of cash) at time t
C0 : the capital investment ( t = 0 ))

The table below illustrates a fictitious project utilizing TCO. In this example, the project
investment is 100 k€, and maintenance is expected in years 5, 10 and 15. The technical
lifetime of the system is 20 years.

Table 4: Fictive project, TCO calculation

TCO Expected lifetime 20 year

Cost Year 1 Year 5 Year 10 Year 15
Investment Maintenance Maintenance Maintenance
Type of cost
C0 Ct Ct Ct
Expected cost 100 k€ 5,00 k€ 2,00 k€ 5 k€
100 + 5 + 2 + 5
TCO = 5,60 k€/y
20

A major disadvantage of the TCO calculations is that the value of money over time is not
taken into account. In this example, it is supposed that the value of money in year 15 is
the same as today’s money. In other words, inflation and interest rate are not taken into
account. Another drawback of TCO calculation is that the benefits are ignored.

25/80
4.2.2 Net Present Value

Net Present Value (NPV) is a financial tool whereby it is possible to compare different
projects with each other. The basis of NPV is that value of future cost/benefits is
calculated back in today’s money. For calculating back a defined interest rate percentage
is used. This high value of the interest rate is in most cases defined by the regulator or at
the corporate level. An often-used interest rate percentage in a NPC calculation is the
Weighted Average Cost of Capital (WACC).

With the same formula also benefits can be calculated. The NPV methodology can be
combined with other financial models.

Equation 3: NPV
n
Ct
NPV = ∑ (1 + r ) t
− C0
t =1

Where:
t: the time of the cash flow
n: the total time of the project
r: the discount rate
Ct : the net cash flow (the amount of cash) at time t (positive benefits, negative cost)
C0 : the capital investment ( t = 0 )

Remark: in this equation the cost is calculated as a negative value and benefits as a
positive value.

Table 5 is for the same fiction project using the NPV calculated.

26/80
Table 5: Fiction project, NPV calculation

Expected cost Expected NPV

benefits 6% interest
Year rate
0 Co -€ 100.000 €0
1 Ct € 10.000 € 10.000
2 Ct € 10.000 € 9.434
3 Ct € 10.000 € 8.900
4 Ct € 10.000 € 8.396
5 Ct -€ 5.000 € 10.000 € 11.881
6 Ct € 10.000 € 7.473
7 Ct € 10.000 € 7.050
8 Ct € 10.000 € 6.651
9 Ct € 10.000 € 6.274
10 Ct -€ 5.000 € 10.000 € 8.878
11 Ct € 10.000 € 5.584
12 Ct € 10.000 € 5.268
13 Ct € 10.000 € 4.970
14 Ct € 10.000 6.635 €
15 Ct -€ 5.000 10.000 € € 4.688

NPV € 12.081

Only those projects that have a positive NPV are worth investing in. Those projects that
have a negative NPC should be in principle rejected.

For secondary systems is it very hard to calculate the real benefits. Sometimes utilities
choose to calculate only the NPV cost. They assume that the benefits for the secondary
systems are the same.

4.2.3 Internal Rate of Return

Internal Rate of Return (TRR) is a methodology to calculate the interest rate for a project
whereby the NPV is zero. It is important is that the cost and benefits are known. The same
NPV equations are used to calculate the IRR. The difference with the NPV is that the
interest rate is a given figure and in an IRR calculation the interest rate is calculated. The
advantage of this method is that it is not necessary to know the interest rate.

Equation 3: IRR

n
Ct
Initial Investment = ∑ (1 +
t =1 IRR )t

27/80
Where:
t : the time of the cash flow
n : the total time of the project
IRR: internal rate of return
Ct : the net cash flow (the amount of cash) at time t (positive benefits, negative cost)
C0 : the capital investment ( t = 0 )

Figure 3 is an example of IRR calculation. In this example the NPV is calculated with
different interest rates. The IRR is the interest percentage whereby the NPV is zero. In
this example the IRR is 17.5%.

€ 60,00

€ 50,00

€ 40,00

€ 30,00
NPV *1000

IRR
€ 20,00

€ 10,00

€ 0,00
0,00% 4,00% 8,00% 12,00% 16,00% 20,00% 24,00%
€ 10,00-

€ 20,00-
Interest rate %

Figure 3: Example IRR calculation

Financial experts are mainly using this tool. The IRR tool is basically not intended to
compare different projects with each other. The idea behind this tool is that financial
expects can decide if a single project is valuable to invest in (only project with a IRR value
lower than the interest rate are worth to invest in). A drawback of this tool is that also the
benefits should be known.

4.2.4 Return on Investment

The Return on Investment (ROI) or Ratio of Return (RR) is the ratio between investments
and the direct accountable benefits. In a ROI calculation the interest ratio is ignored. It is
also possible to combine a ROI calculation with net present values. Besides the ROI
calculations with NPV there are other way’s to calculate the ROI. In principle the ROI

28/80
equation is very simple. The drawback of a ROI calculation is that for protection and
control systems it is very difficult to calculate benefits. Normally the period for ROI
calculation is over a specific period of time, usually a year. In most cases is the ROI
calculation not suitable to compare different projects with each other.

Equation 4: ROI

Vb − Vc
ROI = (%)
Vc

Where:
Vb: Benefits accountable to the investment
Vc: Cost

Other tools such as TCO or NPV give better results. In the standard equation inflation and
interest rate are ignored. Normally the ROI is a figure that can be found in annual reports.

4.2.5 Payback Period

Payback Period is a methodology to calculate the period of time that is needed to earn the
investment back. In a standard Payback calculation inflation and interest rate are ignored.
A Payback calculation can be combined with net present values.

V investment
Payback =
V cashflow ( year )

Where:
Vinvestment: Investment cost
Vcashflow: Cost minus benefits

For example with an investment of 100 k€ and a yearly benefit of 25 k€ it will take 4 years
to have hit the a break even point. In this example the payback period is 4 years.

A Payback calculation gives insight into the time frame that is necessary to earn the
investment back. Other calculations models gives in many cases a more balanced result.

4.2.6 Life cycle cost

Figures in TCO calculations are usually based on the real expected cash flow. In a NPV
calculation expected cash flow is calculated back into today money. See chapters 5.2.1
and 5.2.2. Life cycle cost calculation is a combination of TCO and NPV calculation. Cash

29/80
flow in a LCC calculation is not presented in real expected money but based on a NPV
calculation. The time frame for a LCC calculation should be long enough that at least all
systems involved in the calculation are replaced. Typical timeframe for a LCC calculation
is between 20 and 40 years.

In a LCC calculation it is possible to add risk and benefits as cost in the LCC calculation.
Risk and cost shall be calculated in Euro/year. The risk and benefits can be put in the
calculations as yearly cost or benefit.

30/80
4.3 Summary

Equation Positive Negative

TCO t =n
− All costs are calculated − Benefits are not taken into
∑C 0 + Ct
TCO = t =1 − Calculation period can be account
n long − Interest rate and inflation
− Similar projects can be are ignored
compared
NPV n
Ct − Interest rate and inflation is − To determine the interest
NPV = ∑ (1 + r ) t
− C0
t =1 part of the calculation rate is not always simple
− NPV calculation can be used − Not the real cash flow is
with other tools not calculated
− For a protection and
control system it is difficult
to calculate the benefits
IRR n
Ct − The value of the interest rate − This tool is not intend to
IRR = ∑ (1 +
t =1 IRR )t is not needed compare different projects
− Interest rate and inflation is with each other
part of the calculation − Not the real cash flow is
not calculated
− For a protection and
control system it is difficult
to calculate the benefits
ROI Vb − Vc − Equation is very simple − Interest rate and inflation
ROI =
Vc − ROI is calculated over a are ignored
specific period of time, usually − Other tools such as TCO
a year and NPV gives a better
result
− For a protection and
control system it is difficult
to calculate the benefits
Payback V investment − Equation is very simple − Interest rate and inflation
Payback =
V cashflow ( year ) are ignored
− Other tools such as TCO
and NPV gives a better
result
− Difficult to calculate the
cash flow
LCC Combination of TOC and − Gives best and most −
NPV objectives results
− The risks and benefits
can be taken into
account

31/80
4.4 Examples

4.4.1 Introduction

In this example, operation risk and financial considerations are analysed in an objective
way.

In this example two systems are compared with each other, namely:
system A: that is an existing system in operation for many years. There are technical
constrains in this system. For example, the knowledge of how to maintain
this system is not in the company and the support of the supplier is below
an accepted level.

system B: a new system with new functionalities. The new functionality has many
benefits, for example remote diagnosis, extensive self-supervision
functionality.

System A: Existing system System B: New system

Investment cost (CAPEX) Investment cost (CAPEX)
Operational cost (OPEX) Operational cost (OPEX)
Risk (high) Risk (low)
Benefit (low) Benefit (high)

Figure 4: system A and B what to do?

The driving force to start a study might be that risk of system a is too high or that it seems
that system B has so many benefits that it is recommended to replace system A. Both
trigger items points are described in to different examples.

Example 1: system A is in operation and there are problems with this system.
Question, is it from a risk and cost point of view justified to replace this
system with a new one with the same functionality?

Example 2: is a new system available in the market, this system has many benefits.
Question, is it from cost and benefit point of view justified to replace the
existing system before the end of its technical life?

32/80
4.4.2 Example 1: high risk of existing system

In example 1 is an existing system that gives a lot of problems. Advisee is to follow the
following flowchart, see Figure 5.

System A: Existing system

Investment cost CAPEX
Operational cost OPEX
Risk (high)
Benefit(low)

LCC calculation
Compare OPEX No replacement
and CAPEX LCC calculation continue with system a
between Year cost system with modification if
Risk assessment system A and new A > system B
Identification system B applicable
Assessment
Environmental
Safety
Regulator
Stakeholders
Potential risk treatments LCC calculation
Accepting the risk Compare OPEX
Migration and CAPEX the
Elimination cost of the
modification
between system A
and new B

System B: New system

Replace system A Investment cost CAPEX
Operational cost OPEX
for new system B Risk (low)
Benefit(high)

Figure 5: flowchart example 1

33/80
Starting point is an existing system A.
The first recommended step is to make a risk assessment. In this risk assessment all risk
are identified and compared with the company values. Output of this risk assessment is:
− risk is acceptable, no additional steps are needed
− risk is higher than accepted, but with some additional actions the risk can be
controlled (lower than maximal accepted by the stakeholders)
− the risk level is too high and the system should be replaced.

The next step in the study is a LCC calculation. If the outcome of the risk assessment is
that the system should be replaced (c) it is in principle not necessary to make a LCC
calculation. If the outcome of the risk assessment is that no additional steps are needed
(a) or with some modification in the system (b) the risk is decreased and lower than
maximal accepted a LLC calculation shall be performed.

If the outcome is that no additional steps are needed is it justified to make a LCC
calculation with only the yearly operational cost (OPerational EXpenditures OPEX) and
the yearly investment cost (CAPital EXpenditures CAPEX) In this calculation risk and
benefits are not taken into account. Based on the outcome of the calculations it might be
still possible that from cost point of view advisable to replace system A for system B

If the outcome is that with additional steps the risk is controlled a same LCC calculation
shall be made. In this LCC calculation also the cost for the modification shall be included.

4.4.3 Example 2: high benefits new system

Example 2 is a new system with many benefits. It is advisable to assess the benefits of
the new system and analyse those benefits from a cost point of view. The best approach
in this example is to make a complete LCC calculation. The benefits can be part of the
LCC calculations. For example due to improved self supervision the chance of an
unwanted outage is reduced. The yearly benefits can be calculated by the value of the
benefit divided by amount of years this new system is in operation in the network.

In the table below an LCC calculation made for system A and the new system B. In this
example it is supposed that system A is replaced in Year 2 and that benefits of this new
system also start at Year 2. Further it is supposed that the economical debit of system A
stops in Year 3. See Table 6.

34/80
Table 6: example LCC calculation with benefits
Year Scenario A Scenario B
Without replacing system A Replacing system A in year 2 for a new
system
0 Opex cost /year Opex cost system A /year
Capex cost /year Capex cost system A /year
1 Opex cost /year Opex cost system A /year
Capex cost /year Capex cost system A /year
2 Opex cost /year Switching cost
Capex cost /year Opex cost system B /year
Capex cost system B /year
1
3 Opex cost /year Opex cost system B /year
Capex cost system B /year
Benefits system B /year
.. Opex cost /year Opex cost system B /year
Capex cost system B /year
Benefits system B /year
20 Opex cost /year Opex cost system B /year
Capex cost system B /year
Benefits system B /year
∑ NPV ∑ NPV

If the LCC shows that the overall net present cost of system B is lower than system A than
it is from cost point of view is it justified to replace system A before the end of the
technical life time. In the LCC is most difficult point is that the expected benefits are in
most cases not easy quantifiable.

4.4.4 Overall methodology

Both scenarios can be combined. Also the risk can be translated to cost since risk is
change multiplied by effect (the effect can be translated into EURO’s). For a LCC
calculation the risk and benefits are treated in the same way as OPEX, CAPEX. The
difference between risk and benefits is that benefits can be abstracted off the year cost
and risk is added cost.

1
Remark: in this example the CAPEX of system A stops due to end economical investment debit
system A

35/80
Table 7: overall LCC calculation
System Risk Cost Benefits Total
A: Existing High Investment cost (CAPEX) ↓ Low
system Operational cost (OPEX) ↑

€ /year ↑ € /year ↓ - € /year ↓ ∑ € /year

B: New Low Investment cost (CAPEX) ↑ High

system Operational cost (OPEX) ↓

€ /year ↓ € /year ↑ - € /year ↑ ∑ € /year

36/80
5 OPTIONS AND STRATEGIES FOR CHANGE

This chapter is dealing with migration possibilities, general methods and challenges in this
process. It is ending with actual experiences from United Kingdom, Finland, Poland,
France and Belgium.

5.1 General

It is widely accepted within the industry that most of the installed protection relays are
rapidly approaching or have passed their operating lifetime. According to Newton-Evans’
‘Worldwide Study of Protective Relay Marketplace in Electric Utilities’ approximately 60%
of the relays currently being in operation are electromechanical ones with age of 30-40
years. This number varies significantly from country to country. It is recognised that there
is a need for management of the obsolescence of components of such aging devices.
Several methodologies are currently employed to achieve this goal, such as: relay
replacement, refurbishment and hardware life cycle management.
A number of possibilities and techniques to improve the transition from aging protection
systems to up-to-date modern technology with minimal disruption to the power system are
available, namely: maintenance contracts, protection upgrades, spares holdings,
refurbishment of existing infrastructure and life cycle monitoring.

5.2 Migration Options

As with all Engineering solutions, it is important to consider the difficulty of the task and
the resulting technical / economical advantages. If possible it is better to keep it simple to
reduce the design and testing time. However, once it is determined that the current
installed equipment is in need of replacement, and then various options are available.
These range from `plug and play` refurbishment kits to complete scheme upgrade and
replacement. These options will require, to some extent, a combination of additional
competences, including full short circuit analysis, protection setting studies/examination,
programmable scheme logic design, drawing layouts including AC/DC schematics,
understanding of current technologies and protection scheme, commissioning, training
and comprehensive project management.
This approach provides the opportunity to modify, improve and update the protection
system through the use of modern technology. The following available options depend
wholly on utility requirements; primarily, capital expenditure, utility history and legacy
practices, system down time availability, complexity or simplicity of the scheme as well as
personnel skills and human assets.

37/80
‘Plug and play’ Kits
This involves retrofitting modern protection equipment in existing housing. This has the
advantage of no panel rework with minimal scheme wiring changes, resulting in
economical capital cost expenditure and minimal system down time. Associated additional
costs, such as design work, Computer Aided Design (CAD) and project management are
minimised. However, this is applicable only for very small changes, i.e. replacement of a
single relay or similar. Otherwise, the work will become more costly than expected.

Panel Door or Rack Replacement

This methodology enables creation of the pre-wired and pre-tested protection scheme
remote from the final installation site. This minimises system down time, since the whole
panel is replaced simplifying the installation and commissioning stage. Such designs can
take the form of complete panel doors, panel sections or cut outs and various rack
arrangements. However, this requires extra work at the factory for testing when the wiring
goes via other parts of the cubicle that are not present at factory. And all the wiring
between panel/rack must then be added at site. Only if the whole panel or cubicle is
constructed, full advantage can be achieved regarding pre-tested circuits.

Extension of Existing Installations

This option enables modification to existing relay panels or the addition to existing relay
panels. Such installations can include any combination of the designs detailed above, as
well as complete panels to expand or modify existing protection schemes and substations.

Complete Scheme Replacement

Complete scheme replacement may result from the fact that the existing protection is no
longer adequate or where modification is uneconomical. This is the most expensive
solution in terms of initial capital expenditure and system down time. However, it provides
the best platform for taking advantage of modern technology.

Protection System Replacement

Often when a protection system is refurbished, economical restraints prevent the
associated switchgear from been updated. This is not an ideal situation since at some
point the switchgear will have to be replaced, resulting in additional system outages.
Remark: some companies prefer to wait until change of primary and secondary part is
possible at the same time (e.g. due to limited budget). Reason is to minimize the impact
on the services provided to customers.

Control System Replacement

In case of need of improvement of quality and operation measurement data and binary
data for SCADA, EMS, equipment supervision, it is sometimes more economic to replace
the old RTU or control system than to modify the existing equipment.

38/80
Protection Relay Component Replacement
Provided that aging components are identified, available on the market and easy to
replace this method is an economic solution.

When migrating from legacy to new systems following issues need to be addressed:

Compatibility of new generation relays to old type panel layouts and operability with old
generation trip coils/ lock out relays etc.

Review of electromagnetic environment and ambient condition (high level of dust and
humidity)

Wiring and termination practices

DC power supply system, level of DC transients due to switchgear operation, etc. and
grounding arrangements.

Since refurbishment process is implemented over a longer period of time, devices of

multiple suppliers can be confusing due to non standard way of information presented to
operating staff.

Extensive familiarization/ training required to protection and setting engineers for relay
configuration, understanding of features etc.

Management of relays software tools and versions along with accessories

Relay setting data management and adoption of settings.

Chemical and environmental pollution

Breaking and closing capacity of solid state relays (SSR) – dI/dt, dV/dt

39/80
6 UTILITY FEEDBACK – APPROACH & PRACTICES

6.1 Summary on Questionnaire

Due to the variety of practices employed worldwide, the working group deemed it
necessary to consult a wider population. Hence, a questionnaire was prepared and
distributed to a number of utilities: Its goal was to capture their experiences in the area of
secondary system refurbishment. The main areas under consideration were as follows:

• Criteria used in the decision process

• Refurbishment strategy
• Migration Strategy
• Costing Strategy

In all 12 utilities responded. The results have been consolidated and detailed below:

6.1.1 Criteria used in the decision process

The utilities were provided with a list of criteria and asked to name the ones that
influenced the decision process. They were then further asked to rate them in order of
importance. Table 8 details the results: a score of 10 indicating the highest influence.

Table 8: Results: criteria used in the decision process

10
9
8
7
6
5
4
3
2
1
0
Extensibility
Investment

Improv protect

Service support
Manuf suppo
Quality

Manuf chang

Remo main

Reliability

Maintenance

Dependability
Reduce maint cost

Outage cost
Inform & int

Lack educat

Communicat

Fault locat
Oper risk

lack doc

Add functions
Spares
Person risk

Prog tools

Licenses
Warranty

40/80
From the results detailed in the table it can be seen that the following issues play an
important role in the decision process.

Operational Risk
Continuity of supply is an important consideration for all utilities. Ageing protection
systems may present a number of operation risk issues, such as; maloperation or even no
operation. It is very important to access the level of risk associated with the current
installed base. When investigating operational risk you need to consider reliability, security
and dependability.

Manufacturers Support
As protection relays become more technologically advanced, with increasing features,
settings and functionality; utilities are becoming increasingly reliant upon the support of
the manufactures. This can take the form of after sales service or training.

Quality
The quality of the protection system scores highly on the list. It is important to understand
that this extends beyond the physical system itself, it is necessary to have a good quality
management system, i.e. document, setting and drawing control along with a robust audit
system

Investment / Reduced Maintenance Cost

The economical drivers should always be considered along with the technical aspects to
ensure that an acceptable balance exists. Reduction in both maintenance costs and
outage times brings considerable financial rewards and reduces operational risks (less
down time).

Lack of Documentation
Lack of documentation is a major issue for aging systems. Documents are often
misplaced, damaged or not kept up to date. Without such documentation it is very difficult
to maintain a system and prolong its life.

Improved Protection (Protective Relay Functionality)

Advances in technology allow protection systems to be upgraded and other functions such
as event, disturbance recording and self monitoring to be employed which may have been
prohibitive when the original protection system was implemented.

41/80
Service Support
It is important to ensure that if a system fails service support is available. This becomes
more difficult as the system ages, since experience in such systems may deplete over the
years.

6.1.2 Strategies

Refurbishment Strategy
All utilities that responded agreed that the number one driver behind refurbishment
strategy is the condition of the currently installed equipment; based on a condition
assessment. However, there was considerable variation amongst the Utilities polled as to
what level they plan to refurbish at. The following areas were mentioned: network level
(many substation), substation level (limited number of substations), bay level, circuit level,
relay level (individual relays within a bay).

Migration Strategy
In general the migration strategy can occur at all levels, namely, Sub, Bay, Circuit. It tends
to be progressive in nature and driven by the age and condition of the installed base.

Costing Strategy
When polled on costing strategy, all except 2 utilities consider only the life cycle costs, not
the initial costs. The two exceptions take into account NPV. When asked “Do you consider
savings on account of combining various functionalities in a single numerical device,
reducing space/wiring etc to justify higher costs”, the overall response was yes, however,
it is voltage dependent.

6.2 Practical Experiences in Some Countries

6.2.1 United Kingdom approach

Case 1 – Rail Industry

It was noted that a considerable amount of distance protection relays were fast
approaching replacement for the following reasons:
− Relays were obsolete and no longer supported by the manufacturer
− The Rail Industry had a limited amount of spares
− Current protection scheme unable to deal with current demands i.e. temperature
monitoring and regenerative breaking.
− Remote control and monitoring was required
Due to the strategic nature of the rail system and the requirement to reduce outage times
to a minimum, a simple and fast solution was required. The current protection schemes

42/80
were on hinged doors. To reduce the amount of time on site, a complete new pre-wired
door, with flying leads for external connections was built in the factory and tested before
delivery to site. It was then possible to simply remove the existing door and wiring to the
terminal blocks in the back of the panel and replace with the new panel and test within
one day. To speed up any interpanel wiring, cabling between panels was run in but not
landed while the system was live, awaiting connection during the outage. This pre
enabling works dramatically reduces the system down time.

Case 2 – Distribution Network Operator (DNO)

In order to move with technology a number of DNO`s are embracing numerical relays and
taking advantage of the features now available within such devices. This is often a staged
process to gain confidence with the following most popular combinations in use:
− Main Protection only, rest of the scheme maintained
− Combine auto-reclose protection into numerical main protection and discard auto-
reclose panel.
− Replace all, but have two identical numerical relays, main and back-up
− Complete replacement
Where a considerable amount of relays are to be replaced, then a new front sheet is the
preferred solution, since it reduces the outage time. Such schemes can be tested in the
factory to ensure the scheme is functioning correctly before delivery to site. Once on site,
it is then necessary to test the interface to the system.
To date, the majority of the refurbishment projects have been protection panels only and
no switchgear. This is often due to economical constraints. However, in the future this will
probably change and complete systems will be replaced i.e. protection plus switchgear.

6.2.2 Finland approach

It seems that some electronic (static) relays and first generation microprocessor relays will
need extra maintenance or maintenance in advance. The ageing of some electronic
components of these relays is faster than expected. For example, some capacitors,
potentiometers and trip relays are not so old. Our company has had 5-6 false trips during
the last two years (in medium voltage level) from electronic and first generation numerical
relays. It seems that the coverage of the self detection is not very good in these (late
1980's early 1990's) first numerical relays. We have also considered that in the case of an
internal error, false trip of the relay is the last possible action. This seems not to be true
necessarily. Hopefully, the internal fault detection function of the latest generation
numerical relays is as good as manufacturers advertise [claim].
Solution: The manufacturer of the relays has proposed that our company (and the others
also in Finland) should start to do preventive maintenance actions: Risky parts of the
relays should be changed with the help of repair kits provided by manufacturer (risky

43/80
electronic card types, bad capacitors, set potentiometers, trip relays, etc.). It is more
economical to do this at the same time of periodical relay testing in the substation. The
cost of this job should be about 10 % of the cost of changing the relay to new one. With
this they promise 10 years "good" lifetime after the event. This action is reasonable to do
for relays which are about 10 to 20+ years old. This preventive maintenance seems to be
a reasonable solution at least in Finland, because we have a large amount of electronic
and first digital relays assembled in 1970’s and 1980's and these devices are made by the
same manufacturer. The total change to new relays is impossible in these scales. Other
manufacturers and countries also have met this problem.

6.2.3 Poland approach

In 2003 PSE-Operator SA (TSO in Poland) established a new project called ‘Remote

Control and Supervision of the Transmission Grid’. This is aimed at introducing fully
remote operation of the transmission grid in Poland by implementation of the state-of-the-
art substation automation systems throughout the network and implementation of a new
organization scheme in the areas of power system operation and transmission assets
management. This process should lead to unmanned operation of the substations
performed by remote control and remote supervision from the centres.
There has been agreed that the primary driving force for this project is common and direct
access to the information required by each individual organization unit inside PSE-
Operator.
The scope of individual projects (substations’ refurbishment) varies and depends on the
actual condition of the secondary as well as primary equipment. Beneath are described
the most representative examples.

Case 1 – substation replacement

Installation of the state-of-the-art Substation Automation System was included in the
scope of comprehensive modernization of a substation. It assures full implementation of
current standard functional specifications for substation and overhead HV/EHV lines.
Those standard specifications have been prepared by the PSE-Operator’s technical staff,
published on the web site, and are mandatory for all equipment and services providers.
Such a global refurbishment makes it probable to ‘forget’ the problems with functioning of
certain pieces of equipment and substation as a whole, providing decrease of
maintenance costs and appropriate environment for the operation of the grid in a new
organization scheme.
Usually it concerns running substations with energized circuits. The upgrading process
has to provide for an uninterruptible supply to the vital loads. It is provided by applying
step by step installation and rolling shut downs of the consecutive bays. It requires from
both provider and customer preparation of ‘schedule of work’ by using an appropriate

44/80
method and form, e.g. CPM and Gantt chart, which has to be agreed on by Asset
Management Department and Grid Operation Department, especially regarding shut
downs of HV lines and power transformers.
As an example, modernization of Mory 220/110 kV substation located in Warsaw suburb
area, which is the essential supply node for Warsaw agglomeration. The substation was
first energized in the mid 1950’s. The primary as well as secondary equipment and
substation infrastructure have reached the end of their lifetime and therefore the
substation needed a deep refurbishment. One could say that the old substation has been
completely destroyed and at the same place a new substation has been built.
The task the contractor was faced with was a real challenge – to perform a substation
replacement with minimal shutdowns. To fulfil the above requirements the whole project
has been divided into six phases. It started in August 1999 and finished in September
2004.
Modernization has been carried out by keeping the substation’s scheme with slight
changes to its default operating schedule and there was no need to introduce changes in
grid topology, especially highly inconvenient T-connections in the front of the substation.

Figure 6. View of 110 kV switchyard

The substation now consists of two switchyards – 220 kV with 15 bays and 110 kV with 25
bays, each switchyard equipped with triple busbar system, three power transformers
220/110 kV, 160 MVA each, auxiliary supply switchgears AC and DC (see Fig.6).

45/80
The upgrade process has resulted in installation of new primary and secondary
equipment, like:
220 kV and 110 kV circuit breakers,
− Instrument transformers (voltage, current and combined transformers),
− disconnectors,
− digital protection terminals supporting IEC 60870-5-103 protocol (see Fig.7),
− substation control system supporting all nowadays required functions, including
bidirectional communication with control centres, protection terminals and other
IED’s in the substation (see Fig.7),
− SDH communication transmission equipment supporting voice and data channels,
and transmission of teleprotection commands

46/80
 −

Figure 7. Protection and Control cubicles layout

Case 2 – substation retrofit

Due to limited funds we installed only specific components of SAS. The scope of the SAS
implementation depends on the technical condition of substation components, both
primary and secondary circuits. Usually it includes installation of substation control system
and replacement of the most exploited and vital components of the protection scheme,
e.g. busbar protection breaker failure protection. Sometimes it also involves minor
changes in (or retrofit of) primary devices, e.g. disconnector drives. In such a case the
focus is on integration of the new components with the remaining secondary and primary
devices.
An example of such an approach is the 220/110 kV shared substation. ‘Shared’ means
that the 220 kV switchyard including two power transformers 220/110 kV and their 110 kV

47/80
bays are owned by PPGC, while the rest of the substation including 110 kV switchyard,
auxiliary supply switchgears (0.4 kV AC and 220 V DC) and substation buildings are
owned by distribution company.
On the distribution side there are still static protection relays, there have been done some
refurbishment of marshalling panels and control panels (providing local control and
measurements). A new digital control system in a centralized architecture supporting DNP
3.0 transmission protocol for external communication has been installed
On the transmission side, electromechanical distance protection relays (used as main
protection) have not been touched. The replacement covered electromechanical earth
fault relays (used as backup protection) and old TTL-technology based RTU. Digital earth
fault relays supporting IEC 60870-5-103 communication protocol and digital substation
control and supervision (SCS) system in distributed architecture supporting DNP 3.0
protocol, used for communication with SCADA systems in control centres and IEC 60870-
5-103 protocol for communication with protection IED’s have been installed. The
communication features of the SCS included also implementation of TCP/IP stack,
IEC 60870-5-104 protocol and Web Server application to allow communication via
enterprise WAN.
On figures 8 – 10 are presented some typical screens of new SCS and current SAS
communication scheme. This retrofit process allowed for improved substation
observability, quality of measurements and topology information transmitted to SCADA
systems. Thanks to the implementation of remote control it was possible to resign from
the local manned operation of the substation – service provided by a distribution
company.

48/80
Figure 8: Substation layout on the local HMI

Figure 9: Detailed layout of the feeder incl. protection status, bay measurements, etc.

49/80
Figure 10: Supervision on data transmission in the Substation Automation System

Certain projects differ depending on the approach to bidding and ordering process. From
the view of „openness” there can be public bids – open to all vendors, and limited bids –
directed to pre-selected vendors. From the view of completion of projects there are
contracts which scope covers certain part or parts of the substation and turn-key contracts
covering all works in the substation.

6.2.4 France approach

Is it really necessary and urgent to refurbish / retrofit / replace?

This question arises when the protection relays cause incidents, either by failure on
request, or by spurious operation on external fault or even without High Voltage cause.
The answer is easy for the old electromechanical equipment, whose measurement relays
are not manufactured any more. It is less simple for the analogue electronic equipment for
which it is still rather easy to find components for the repairing: indeed, we can wonder
whether the equipment is at the end of its lifetime or if its reliability remains sufficient in
spite of the incident.
Described below is the method we use which allows us to say which equipment could still

50/80
remain in operation for 10 to 20 years plus, on the condition of carrying out some
maintenance actions at regular intervals.
To achieve this goal, a protection of each
type was taken out of the network for
thorough analysis.
We started by measuring directly on the
charts the characteristics of the
components for which ageing phenomena
are known.
The aluminium electrolytic capacitors are
easily measured without disassembling
the chart (measurement at 1 kHz, with a
bias voltage of 2V), because their
Figure 11: Capacity variation capacities are high and they have a low
impedance versus the other components
of the chart (of course it must be checked on the chart scheme that there is no other
capacitor in parallel). Thus we confirmed the ageing of these components (the smaller
they are, the more they age quickly): this led to the recommendation of replacement every
12 years for the values lower or equal to 100µF and every 18 years for the others. We
also observed that some capacitor manufacturers had products ageing much more quickly
than the average: we recommended their replacement as soon as possible by products of
reliable manufacturers.
The photo couplers may age by reduction
of the light efficiency of the emitting diode.
Measurement on chart of the CTR
(Current Transfer Ratio) is possible by
making some corrections to the currents
of measured input and output: the output
is polarized under a constant voltage of
5V and the current is measured under a
null input current: this current is the
consumption of the other components of
the chart which is constant under a
constant voltage: it is subtracted from the
other measurements. The correction of
the input current is done by analyzing the
Figure 12: CTR% correction
diagram in order to calculate the current
leaking in the other components under the input voltage (the photo couplers being used to
isolate the inputs, the studied diagram is finally very restricted). No sign of ageing was
found, even for those permanently active. The photo coupler manufacturers seem to have
controlled this problem. Recommendation: a new sampling will be carried out in 10 years.

51/80
 The coding wheels present a risk of opening of
their internal contacts: this defect is difficult to
Co Re
characterize because it is often fugitive (it is
sufficient to move the wheel so that it
Ca disappears and then reappears later). Using
the schemes we analysed the effect of such
0 1 2 3 4 5 6 7 8 9
openings on the behaviour of the protection:
measure distorted of x%, spurious trip order on
Figure 13: Coding wheels external fault, pickup thresholds modified. We
recommended the replacement of the coding
wheels when spurious tripping was proved (note: as protections are doubled, the failure is
definitely less critical). It was then necessary to propose a replacement system which
really eliminates this risk: we replaced the fixed part by a connector base plate and the
moving part by a connector block in which a welded wiring reproduce the code (thus there
is as many blocks as figures).
The dip-switches, which usually do not cause big
problems, revealed a series badly manufactured: the
plastic block opens and the contacts also: A
recommendation of immediate replacement was made,
with this difficulty that these components are not dated and
not marked; the help of the equipment manufacturer is
necessary to know the charts manufactured at the same
time as those where we found the problem.

Figure 14: Dip-switches

Then we continued with

measurements on components of
which no ageing mode is known
(in any case under the rather soft
conditions of our equipment): The
transistors were the subject of a
statement of I(V) curves for their
2 junctions: the exponential
characteristic of the junctions Figure 15: Junctions
allowed to neglect the low values
and to measure the variation with
ideal exponential for voltages between 0.6V and 1V. This variation allows to calculate the
series resistance of the junction: each family of transistors presents a rather constant
value, with a value a little higher for the Base-Emitter connection than for the Base-
Collector connection. Some specimens far from the mean value were taken for analysis:
most probably the cause is a manufacturing defect: bondings badly positioned, silicon

52/80
support striped. The diodes were also measured with the same process. We highlighted
that certain batches were residues of sorted batches: characteristics beyond 3 standard
deviations. A study of the schemes and a simulation showed that the correct operation of
the chart remained largely assured in spite of these defects.
Id pour Vdd = 1V
Then we tested the logical
0,1
components, which present
0,08 the characteristic to have a
0,06
I d (A )

0,04
measurable junction (without
0,02
protective series resistance)
0
between the outputs and the
power supply pins: Gaussian
dispersion, except for a batch,
Figure 16: measure
probably also a residue of
sorted batch: then it was checked using the chart scheme that the reduced fan-out
remained compatible with the load. Finally we tested the operational amplifiers, which
present the characteristic to have between the 2 inputs or an input and the GND a
measurable junction (without protective series resistance). Substantial but non Gaussian
dispersion, due to the fact that several inputs in parallel are often measured: by dividing
the current by the number of inputs in parallel, dispersion is then very weak and Gaussian.
The recommendation of
maintenance was to do nothing.
For the repair of broken down
charts, an LM308H operational
amplifier of the Eighties had a
particular characteristic: its
output could be clamped by
Figure 17: scheme
referring the compensating
network by a diode to the wished
limit voltage: certain diagrams use this property which is not available any more on the
recent components: we recommended to make a stock of these components.
In conclusion, with respect to the periodic replacement of the aluminium electrolytic
capacitors and a levelling of certain equipment (coding wheels, dip-switches), it is sure
that our equipment will last still a long time.

53/80
 Life Time

Without
maintenance

With maintenance

0 10 20 30 40

Figure 18: life time

An economic survey which develops the fact of delaying of 10 to 15 years the renewal
reveals a very important profitability of this type of maintenance.
The above described approach shows great benefits for the company, although not all
companies can follow it. It requires a lot of expertise, appropriately equipped laboratories,
etc.

6.2.5 Belgian experience

Elia network includes voltage levels in the range 400kV to 30kV (400kV, 220kV, 150kV,
70kV, 36kV and 30kV). All together, the network is composed of 5000 bays all of which
are equipped with Protection and Control systems (cubicle or rack).

Taking into account a life expectancy of 25 years on newly installed protection system,
one can roughly estimate that 200 protection systems (cubicles) have to be replaced on a
year basis in order to maintain the protection and control systems at a good quality level.
This corresponds to a replacement pace of 4 systems per week.

This is a challenging task which has to be tackled with an efficient refurbishment strategy.

Elia protection and control systems are composed of Electromechanical, Static and
numerical equipments that have been installed through the years. The next picture gives
an idea on the technology repartition per voltage level.

54/80
Figure 19: technology

Protection and control devices have been classified in 8 categories, based on their
technology and on their Service Level Index (see hereunder).
− Electromechanical : EM1, EM2, EM3
− Electronic(static) : ELO1,ELO2,ELO3
− Numerical : DIG1, DIG2

Service Level Index:

1: good quality and still available on the market
2: good quality, not available on the market anymore but spare parts are still
available (stock)
3: bad quality or no spare parts available anymore

A more detailed repartition based on the 8 categories is given in the next picture:

55/80
Figure 20: Service level index

EM2 & 3 equipments are expected to have a lifetime of 30-45 years. Those equipments
have been mainly installed in the years 1966-75 and are now reaching the end of their
lifetime.
In the same way, ELO equipments are expected to have a lifetime of 20-35 years. Those
equipments have known their peak of placement in the years 1975-86 and are also
reaching the end of their lifetime.

This situation combined with the fact that life expectancy of digital equipment is not known
but is expected to be in the range of 20-25 years, has led to a double refurbishment
strategy:

1)Investment projects

Maximum replacement of protection and control systems happens through investment

projects. Each project (network reinforcement, network development) is used as an
opportunity to replace the protection & control equipments which are outdated or which
could jeopardize the quality of the system.
As the goal is to replace a maximum of P&C systems, all aspects of the project have been
optimized in order to reduce time & costs. This includes Engineering, FAT, SAT, uptime,
maintenance aspects.

56/80
Practically, this refurbishment strategy makes use of Highly Standardized and modular
Protection & Control cubicles at bay level.

These cubicles are characterized by:

− No auxiliary relays and maximum self supervision (maintenance reduction) ;
− Minimum number of IED’s to cover all cubicles needs (reduction of spare part
costs) ;
− Minimum wiring (reduction of wiring time) ;
− Strict application of international Standards (reduction of approval procedures) ;
− Universal interface to substation level : classical wiring and IEC 61850 connection
;
− Standardized and validated IED’s programming (parameters, scheme logics)
− Standardized layouts, wirings schemes and documentation.

For efficiency reasons, the engineering of each new cubicle is based on a library of +- 50
pre-engineered and standardized cubicles covering 400kV to 30 kV applications and
requiring a minimum of customization.

2) One-to-One replacements

In the long term, the replacement peak caused by EM and ELO equipments combined
with the reduction of the lifetime of P&C equipments and the pressure on investment
budgets could be so binding that this strategy alone could prove to be insufficient.

Replacement through investment projects is therefore completed by a “one-to-one”

replacement strategy. This strategy is aiming at replacing old equipments (EM3, ELO3) in
order to level out the replacement peak through the years.

The one-to-one replacement strategy is based on a so-called “risk model”. This models
makes use of different information (such as equipment age, substation importance, …)
and provides the maintenance engineer with a list a equipment which have to be replaced
with a priority ranking.

The following picture shows the impact of different one-to-one replacement strategies on
the global quality index of the network.

57/80
Without One-to-One replacements:

Figure 21: historical evaluation

58/80
With one-to-one replacements

Figure 22: historical evaluation with replacement after 2011

TCO shows that we don’t need to buy the cheapest solution, because initial cost is smaller
comparing to total cost. It is worth of using more expensive components in order to reduce
operational and maintenance costs.

6.3 Conclusions

It is clear to see from the questionnaire that the main driver for change is continuity of
supply. This is becoming increasingly important as the supply for electricity increases,
since loss of parts of the network may have disastrous consequences. This condition is
often exacerbated by ageing equipment on the system. It is also clear that even though
new technology and additional functionality in the equipment is not a main driver for
change; utilities consider manufacturers support, quality, reliability and advances in
technology as being important aspects in providing a secure and reliable electrical power
system. However, it is important to understand that such advances in technology may
allow the realisation of secondary system designs that were not possible before, hence
improving the network reliability and security.

59/80
An interesting observation is that it is not possible to clearly define the best strategy for
refurbishment, this if often based on a number of factors such as: voltage level,
geography, customer preference etc. However, it is clear that some form of refurbishment
strategy as oppose to complete replacement has an important role to play when balancing
outage times / operational risks against cost.

60/80
7 TRENDS

This chapter describes new functions or changes in existing functions that can help end-
users in the targets which users are confronted with as a result of system and market
trends. This can be input for protection developers to implement this function in the near
future as well as for end users to help in the specification process.
This chapter is divided into the following three items
− System trends.
− Self-check contribution to reduce maintenance.
− Remote maintenance.

One of the driving forces for replacement of new protection systems can be the
introduction of new technology and/or functionality. This paragraph describes future trends
in protection systems. This paragraph can be helpful for end–users and R&D departments
to focus their development policy. It is our intention to describe trends, and not be
complete.
Technical developments should be mainly driven on changing needs of end users.

7.1 System trends

The following trends in the industry and utilities are recognized:

− increasing network load, networks are operating closer to the technical limits
− unpredicted, changing load flows in the network
− reducing investment and operational cost (total cost of ownership)
− improving reliability
− reducing technical skills within the company (outsourcing of non core activity)
− more information from the status of the network to management, load forecast,…
systems etc
− scale and amount of blackouts is increasing
− flexible networks

7.1.1 Increasing network load

The trend is that electrical infrastructure is operating closer to the technical limits.
Worldwide there are two trends. There are networks where the load increases in such a
way that it is almost impossible to follow the increasing energy consumption, and there
are networks where the growth is not so high but where due to the liberalization
investments in the network are postponed.
A second trend is that due to the increasing load end-users decided to up-rate their

61/80
installed primary components (power transformers, cables, overhead lines). In an up-
rating process not the maximum technical value on the rating plate is used. The
construction of a component is analyzed in detail. It has been found out that in many
cases the maximum technical value of a component is much higher than the maximum
value on the rating plate (although this could have an impact in equipment life).
Higher loads in the network and up rating component have an impact on the protection
system / settings. In networks that are operating much closer to the technical limits the
performance of reliable protection system is essential.

7.1.2 Unpredicted, changing load flows in the network

The amount of decentralized energy produced is increasing. Installing large-scale windmill

parks helps countries to fulfil the agreed Kyoto Protocol targets for the reduction of the
emission of greenhouse gases. Due to more dispersed energy, network configurations,
load flows and short-circuit power are more unpredictable. The conventional applications
of transmission protection schemes and methods of calculating protection settings
become more and more unsuitable.

7.1.3 Reducing investment and operational cost (total cost of ownership)

The users of protection systems are confronted with a pressure to reduce investment and
operational cost. Trend is that manufacturers of protection devices develop one uniform
hardware platform, which is suitable for all protection relays. The next step in this
development is that protection functions are delivered as a software module. This
“protection software module” can be installed in any equipment. The relation between
physical hardware and protection functions as a software module will become more
vague. This product standardization allows reducing investment in spare equipments
and/or modules. Modular designs allow reduction in operation costs as in many cases end
users may perform simple repairs at module level replacement on site minimizing
downtime.

Another trend in reducing maintenance cost is the enhanced self-check functionality. The
trend is that self-check functions are also monitoring major parts of the secondary
equipment. Trip-coils supervision and voltage supervision are typical examples. With new
functionality the complete primary equipment can be monitored. When these functions are
available inspection can be reduced to an absolute minimum.

62/80
7.1.4 Improving reliability

There is an increasing focus on power quality. For a modern society the reliability of the
electrical infrastructure is an essential part of economical growth. Regulators are
introducing quality indexes in their financial structure. The reliability of protection systems
is an essential part of the overall network performance.

7.1.5 Reducing technical skills within the company (outsourcing of non core activity)

From published papers it is known that almost 50% of the unwanted protection operations
are related to human errors. With adaptive protection settings (fuzzy and neural network
technology) using the information from network calculations programs and SCADA
systems settings can be automatically calculated. The human expertise of calculating
protection settings can be translated into setting algorithms

7.1.6 More information from the status of the network to management, load forecast,
systems, etc.

Due to the information society the need for additional network information is increasing.
Due to this additional information needs, detailed information regarding the behaviour of
the network is increasing. Typical examples of addition information are for example the
trend recorders which are incorporated in all of digital protection relays. Due to this
additional information it is easier for a protection engineer to explain the cause of a failure.
The next step is that additional information from protection and substation systems is also
used for management of information, maintenance purposes, load forecasting, etc.

7.1.7 Scale and amount of blackouts is increasing.

Networks are enlarged due to globalization. The coherency between power producing and
power consuming are becoming weaker. Interconnecting connections are more than in the
past needed for the network flows and stability. The recent major blackouts in Canada and
Italy show that tripping off a single transmission line can result finally in a blackout of a
complete region. Fact is that the scale, size and frequency of blackouts is increasing.

7.1.8 Flexible networks

The last example of trend is flexible networks. The situation today is that almost every
network is designed and operated in one way. With more flexible networks it is possible to

63/80
configure a network in such a way that in every situation the most economical
configuration is selected. In theory with flexible networks the most cost benefit solution is
selected. Smart grids using modern technologies may contribute to security in a cost
effective way.

7.2 Self check contribution to reduced maintenance

7.2.1 Self supervision definition

We could define the self supervision, sometimes referred to as self-check or self-test, as
the capability of one device to monitor its own health, reporting this failure to upper levels,
and pre-programmed or spontaneous response in terms of its outputs (i.e. disable
protection outputs).
We will see in the scope of self-supervision some cases where the equipment monitors
something more than its own health. These should not be called self-supervision, but it is
important as a global system point of view as it gives value to the user.

7.2.2 Scope of Self supervision

Modern protections include modular design approach, to minimize down time and
increase availability (through quick problem detection and fast module replacement). This
leads to the need of special tests to warranty the right modules were installed in the right
slot when repairs are done in the field. This is why diagnostics are more oriented to these
modules.

Figure 2: Failures by hardware and software components in percentage.

Change over period 1970’s – 1990’s. [Ref. 4]

64/80
On the other hand, modern digital protections implement most of the functions based on
software, so more attention is paid to software elements check, both software and
hardware. In the next chart it is shown the percentage failures occurred by hardware and
software components, being the 90s figure relatively constant and yet valid for present
days.
Software should then be considered as another module. In fact module that contributes to
higher probability of failure. Therefore, some way of software self-supervision should be
also considered. For this purpose, watchdog hardware built in mechanisms in modern
processors may help. Other high level “watchdog” or “token passing” mechanisms may be
used to complement the latest.
Obviously, the scope of self supervision should include the components with the biggest
failure probability for whichever reason.
Software failure modes include not only manufacturer’s bugs but also user programming
bugs (i.e. when programming flexible logic schemes).
Hardware failure modes include output relays and circuit supervision even though
sometimes failures are due to unavoidable wear out, lighting, overloads, etc. more related
to overall application than to the pure IED design or manufacturing quality.

In the appendix 11.3 we describe some hardware and software components that may be
monitored based in existing products. [Ref. 5]

7.2.3 Coverage

It is important for the user to know how much coverage the self supervision provides. In
many cases it is not practical providing a percentage quantitative number due to lack of
data. In other cases manufacturers may give ratios difficult to show against any known
standard. Therefore, qualitative coverage information should be given with a clear
differentiation of two different states:

Start-up
Typically during start-up tests are done that should be disruptive during runtime.
Modern protection equipments with many built-in functions may require significant start-up
time until protection functions are ready to work. In case of electromechanical and static
relays start-up time is negligible (it is in the range of less than 1 second for many
analogue products), whereas it may become important for digital protections. If functions
are enabled in a stepped approach, manufacturer should declare a table of functions
availability and time. For example, protection is ready after 30 seconds of start up,
communication is ready after 1 minute, etc.

Runtime
These tests are intended to diagnose possible real-time failures due to component aging,

65/80
drift, premature failure, verify memory circuit’s integrity, internal buses connections, etc.
Self supervision is done typically during background time of circuits; therefore they do not
affect relay response. Any effect on product unavailability should be declared. If, for
instance, a complete test of a circuit that takes 1 second is done every day, this should be
declared, even though the effect in availability is quite low (12 parts per million).

7.2.4 False alarms

Three collateral effects of self supervision may be foreseen:

Fail to detect a failure

Self supervision cannot cover 100% of circuit failures. Even for simple functions like a
battery voltage monitor, the circuit that monitor battery may also fail. To minimize the
chances of this effect, monitoring or supervision circuit (or software module) should be
simple and robust. In theory it should have failure rate one or two levels of magnitude
better than the monitored circuit, otherwise it is probably a useless waste of resources.

False alarms
This is probably the worst effect for two reasons. Cost, due to attending something that
was not a real problem and loss of detection system credibility that may lead to disabling
the function, tampering or worse: reaction time in future opportunities. Probability of a
false alarm should be reduced, in the order of 1% false alarms

Disruptive test
Supervision should be done in a way that does not interfere with normal protection
functions. Any effect in delaying or even disabling a trip should be considered.
It should be checked supervision sanity.
For instance, circuits with limited number of operations (i.e. an EEPROM memory with,
say, 10^5 operations) should not be operated beyond its limits (i.e. continuous read and
write test during idle or background time at a rate of 1 per second). Some of these effects
are not visible during static, dynamic or endurance tests, therefore they require a deep
product review.
Prevention principle should reign as in other critical areas (medicine, law, etc.).
The simplicity principle should also be used to improve overall reliability.

7.2.5 Self action (response from the device)

When either type of self-test error occurs, the IED will react automatically in some way.
For instance, it may signal the error via local HMI (LED indicators, display). It may also
indicate such error in an internal variable that may be read via communications.

66/80
Depending on error criticality the protection status may change from regular to a
predefined “fail safe” status.
A minor error should not disable protection. For example, in case a measurement
monitoring device detects measurements out of tolerances, the equipment could use
some default calibration parameters that would reduce accuracy to a level still acceptable
in a protection scheme. This behaviour provides higher availability than other that would
disable completely the protection.
In certain cases, user may bias relay towards security or dependability manually or he
may decide to set the relay to react automatically in one way or another. Due to the
possibility of human error and the slow human response time, automatic mechanisms are
preferred. Leaving decision to an additional setting in a complex IED may increase
unnecessary complexity that drives lower usage reliability, therefore settings are not
recommended for these kind of automatic responses.
In cases where a critical major error is found, it is recommended disabling all responses,
including output contacts, to a secure state, i.e. keep output trip contacts open. When this
occurs, user should be able to detect both locally and remotely, this new “protection
disabled” status. Modern IEDs include a “ready” normally closed contact that will open in
such cases. Normally open contacts are preferred because in case of an auxiliary power
outage, it will also go to that “disabled” status signalling something is wrong in the system.

The manufacturer should clearly assess which is the IED response in case of an error and
recommended user reaction in such cases.

An example of how this description could be done is included in appendix 11.3

7.2.6 Prevent malfunction

It is important to highlight that anything done either automatically or manually, should be

focused to prevent mal function. False trips create major losses in processes and should
be avoided. Self check should contribute to an early detection of failures and, very
important too, prevent product mal function, mainly false trips. One way to implement this
“fail safe” mode is disabling all contact outputs to a secure predefined state. In modern
schemes that cooperate via communications in complex protection schemes, such as a
multi-voting scheme in an Ethernet IEC61850 based system, should consider a “dead
relay” as a non valid vote. Even though this is obvious, the fact that in software logic
schemes there is not a conventional schematic diagram that was available in wired
connections, may lead to this utilization error.
Disabling a protection is converting a valuable protecting device into a useless black box.
Therefore, self check mechanisms should be robust, using safe and clear mechanisms in
simple surveillance schemes either software or hardware.

67/80
In certain digital IED’s there is a digital trend by which the relay works perfectly or disables
all functions. This digital behaviour (either works or not) may not be optimum in complex
products that include many protection functions. For instance, if a distance protection
provides overcurrent backup and a voltage input fails (due an internal product error), it
could be considered the possibility of disabling only voltage related functions, while
preserving the simpler overcurrent backup ones. This “analogue” or continuous behaviour
may combine both security and dependability in a better way, providing maximum
availability.

7.2.7 Analyse signalling what information do we need (for efficient maintenance

policy)

Too little data on a failure event may be a bad situation to react. Too much data may be
even worse, provided that some situations are critical and need immediate user response.
It is said that “you can't see the wood for the trees”. A sound self supervision scheme
should provide enough and relevant data. It should also provide a way to obtain useful
information on failure detected, so that the user quickly identifies what failed and knows
what to do in such case.

Present policy of many users is carrying a complete IED to an installation once a failure is
detected. Unfortunately the many models, many manufacturers and different applications,
may make keeping spare IED’s or even spare parts to fix quickly one failure in the field
impractical.

Manufacturers should provide, as much as possible, standard equipments with modular

design and minimum models variation, especially hardware related. Even though updating
modern equipments, flash memory based, via communications, may seem easy,
sometimes users do not have enough training, laptops, or they do not have the required
files in their PCs. For this reason, it would be also recommended a single piece of
software valid for all models and simple product upgrade procedures. All this takes into
account easy maintenance.

Self supervision should lead to easy failure detection. This should provide enough
information on module failure (either hardware or software), for a quick replacement that
will lead to fix field repair. If the product should be taken to user laboratories or
manufacturer facilities, it should be preferred “draw out” types that leave case and wiring
in place.

For an efficient maintenance, it should be considered the possibility of a remote diagnostic

and possibly remote product fixing via communications. This leads to the “tele-charge”
concept that may increase user productivity avoiding travelling on site, many times to

68/80
remote locations in harsh environments. This nice concept should be complemented with
enough access security to prevent product tampering by hackers or inadvertent
manipulation by authorized users.

Taking into account overall costs, each user should find his own optimum point for
periodic intervals.

For meshed distribution lines with frequent faults occurrence, a reactive approach where
user will service an equipment that failed when detected a wrong behaviour, could be
followed without major problems.
For high responsibility high voltage critical short transmission lines with low frequency of
faults occurrence, that reactive approach would be unacceptable. In many cases,
redundant schemes with different manufacturers, models or protection principles, may
help. Nevertheless, the common mode failures make it still important the periodic testing.
In case a common battery with a single fuse may be a weak point even when complex
hardware or software IEDs protection schemes are used.
Period testing should preferably be done with modern automatic test sets that cover a
significant part of the IED functionality. The trend is to carry out complex and
comprehensive testing during SAT (Site Acceptance Tests), while periodic tests are
limited to simpler tests. This was a good approach in the past when simple, reduced
number of functions equipments was used and testing tools were quite simple. Nowadays,
computer programmable test sets may allow quick and productive comprehensive testing
during both FAT and periodic testing.

Self check supervision should be done:

− Simple, sound and reliable
− Oriented towards increased overall availability
− Described clearly by the manufacturer
− Comprehensive
− Following “fail safe” principle
− Clear information to detect what failed and how to react

7.3 Remote maintenance

7.3.1 Remote setting

When protection relays work normally during normal system faults, it is enough to check
relay operation with the fault information and relay recorded data.
However, in case of the complicated system faults where it is suspected a relay mal-

69/80
operation, we need additional information like SOE (Sequence of events) records,
disturbance records, fault reports, relay settings, etc.

In the case of some relays, the data can only be retrieved at the substation. In case of
remote operation system, we can get the recorded data from a maintenance centre. So
we can do the relay operation analysis faster.

The settings change of protection relay occurs as a result of the change of a system
constant and the reconsideration of the system operation. At some utilities, setting values
are changed on the system change by using a remote operation function from a
dispatching centre. For utilities that have frequent system changes, it is very effective that
engineers don’t have to go to the substations.

When implementing settings changes via the remote operation, the security must be paid
attention not to have unjust access from outside. Special setting change that
accompanies facility fault occurs suddenly regardless of the time, most of them caused by
the natural phenomena such as the earthquake, the heavy rain, the windstorm, the snow
damage. Engineers have to go to the substation through dreadful weather. Therefore, if
the setting change by remote operation can be applied, it is possible to secure the safety
of engineers in addition to fast system restoration.
The technique to judge the degradation tendency of the relays is introduced by comparing
a relative error of measuring current/voltage data. Concretely, in case of dual protection
relaying, the sampled current/voltage data are compared with each other at the same time
and we can manage the degradation tendency of the analogue input part.

In cases when flash memories are used for relay, utility is able to rewrite a sequence. For
example, if utility wanted to change a policy of reclosing for line protection, utility can be
able to rewrite a reclosing logic by using remote personal computer. The maintenance
engineer of utility and the technical engineer of manufacturer go to a substation and repair
software now. However, the engineer of utility could repair software by using remote
operation from a maintenance centre.
The labour will be saved by the decrease of the opportunity to go to a substation both of
the utility and manufacturer. The major challenges are:

That the available hardware allows software upgrades.

The way of testing on-site changes done

To secure the reliability and the security of the communications channel.

70/80
7.3.2 Measures for security

In the communication network, when information is transmitted, it goes via various

courses. Therefore, the bugging of the information, the interpolation, destructions by the
unauthorized network user and so on are worried about. To prevent from an invasion into
these networks, the security measure becomes necessary.
Incidentally, whatever measure is implemented, we can’t get a perfect measure. If LAN
connects with the protection relay electrically, we do not know when to be attacked.
The security measure implementing at the utility that applies remote operation is as
follows.
The remote operation communication network is separated from other communication
network physically and prevented from unauthorized user's accessing unjustly from other
network.
Compulsion control commands lead to mal-operation. As for the control order about a
protection scheme, it doesn't make operation conclude only in remote operation. Only if
both conditions by remote operation command and another command are met, would
actually control the equipment. Even if the remote operation network is invaded, the
unauthorized operation to the protection relay cannot be achieved.
It is effective to implement some security measures. The best security measure must be
taken after considering the cost, the effect and the damage of the measures. Therefore,
the company that adopts remote operation must have clear security policy beforehand.

71/80
8 CONCLUSIONS

In this report we have dealt with the various forms of secondary system change. Several
driving factors for change have been described as well as practical refurbishment
strategies in several countries. Cost constraints and technical factors have a major
influence in these business decisions. The authors have sent a survey that has been
answered by different actors (utilities, consultants, etc.) of several countries. Several
conclusions might be extracted from the experience shared between the authors of this
paper:

Definitions

There is no standard terminology for change (refurbishment, repair, replacement) across

different users. The meaning of these terms is related to the context (i.e. component /
system level). The authors have provided definitions for this paper.

Driving forces for change

In certain cases, one could wonder whether there is or not an urgent need for change.
This report describes ways to quantify such need.
The need to minimize down time makes it interesting sometimes replacing both secondary
and primary systems in the same field service.
Main drivers for change have been identified as quality of service, cost of implementation
and risk associated with change or with no change.
New functionality is rarely identified as one of the major drivers for change by users.
Cost is the driving factor for change as many factors may be eventually translated into
cost. Some clues of how to calculate the overall costs are given in this paper.

Refurbishment strategies

Refurbishment strategies are many times conditioned by the need to continue supplying
electricity to certain users.
There are several methods to calculate total life cycle costs. All of them have a common
ground and take into account the initial acquisition costs as only one component of the
many ones that one has to take into account in electrical projects.
There are numerous strategies for change. They depend of factors such as geography,
voltage level, technology, customer know-how and mix of products installed.
Several countries have described their experiences facing product aging, failures, etc
All migration and general change strategies may represent some risks. New digital
equipments may need to take into account EMC (electromagnetic compatibility) and
environmental conditions that were not an issue in electromechanical technologies.

72/80
Practical experiences

From each experience we may learn best practices, taking into account that not all
practices may be applicable sometimes in all situations due to different contexts.
Manufacturers may play a key part in contributing to a seamless and successful
refurbishment, replacement strategy.

Trends

Self checking modern IED’s capabilities may help to reduce the need of commissioning
and reduce overall maintenance time and cost.
Self check may help to improve installation reliability by detecting early failures that could
derive, if not properly addressed, into major system unavailability.

73/80
9 REFERENCES

1. Distance Protections: What Limits of Use for Heavy Loaded Very High Voltage
Lines? J.M. Grellier, J.L. Chaneliere, FR 306, Madrid Colloquium 2007
2. Protection Relay Life Cycle Management, Christopher Smith, John Wright, Steve
Pickering, March 2008
3. Guidelines for Specification and Evaluation of Substation Automation Systems,
CIGRE WG B5.18, August 2007
4. “Digital relay software quality”, Elaine Y. Wintraub GE Protection and control –
Charles R. Heisking , Associated Power Analyst, Philadelphia, 1993
5. “G60 Generator Management Relay UR series instruction manual”, GEK-106411A
version 3.3X, GE Multilin, March 2003. Latest manual version available on the link:
http://www.gedigitalenergy.com/app/ViewFiles.aspx?prod=g60&type=3

74/80
10 APPENDIX

10.1 Maintenance strategy. Example 1.

Beneath there is given an example of maintenance strategy applied by utility operating

transmission and distribution network.
1. guarantee tests before 2 years of usage
2. 3 years periodical testing for electromechanical relays, for electronic (static) relays
and for the first version of numerical relays
3. 6 years periodical testing for new numerical relays with coverable self supervision
function.
4. yearly measurements of internal electronic voltage levels for electronic relays
without self supervision
5. preventive maintenance (replacing risky parts and cards with new ones) for some
electronic and first generation of numerical relays after 15-20 years of lifetime

From manufactory point of view six years test interval is quite long for the new digital
relays. The end-user cannot rely that the internal self-supervision is able to detect all
major internal faults. With the first and the second generation of the numerical relays the
self-supervision could detect not more than 80 % of internal faults. The self-supervision of
modern generation numerical devices is proved; statistics of the self-supervision is not
available. However experiences with internal self-supervision of the latest generation are
encouraging.

10.2 Maintenance strategy. Example 2.

As a general comment, one of the domination parts of the operating costs is the
maintenance and all relating costs. In order to optimize the service strategy, all relevant
cost portions have to be considered. Different maintenance strategies such as time
based, corrective, condition based or reliability centred maintenance have different impact
on life cycle cost. The decisive parameters have to identify and their influence on the
overall cost has to determine, so individual calculations are necessary for the practical
application of these principles. The duly investigation on the life cycle cost and a tight
asset management reduces the overall cost and contributes to the economical benefit of
the operating utility.

75/80
 Basic Organization Chart

MONITORING
INFORMATION

MAINTENANCE ON-CALL
MANAGEMENT TEAMS

MAINTENANCE RENEWAL AND CORRECTIVE

PROGRAMMES REFURBISHMENT MAINTENANCE
PLANS

10.2.1 Example of maintenance measures

Next, example of maintenance measures for protection systems determined by the used
technology is described:

ELECTROMECHANICAL AND ELECTRONIC ANALOG SYSTEMS

Checking of operating characteristics
Operating levels
Response times
Checking of operating logics
Interaction with operating components
Interaction with other equipment
Records control
Checking of the signalling
To the control centre through the RTU
To external event and signals recorders
To mimic board and alarm panel
Checking of established settings
Checking based on theoretical setting list

ELECTRONIC DIGITAL SYSTEMS

Checking of operating characteristics related to System
Operating logics
Operating levels
Response times

76/80
Control of operating logics
Interaction with operating components
Interaction with other equipment
Records control
Oscillographs collected and checking of recorded signals
Control of the analogue-digital converter response
Measurement of in-service values
Measurement under controlled injections (inputs)
Control of established settings
Checking based on theoretical setting list

Furthermore, other secondary system maintenance areas can be:

REMOTE CONTROL UNIT DIAGNOSIS

Inquiry for self-diagnosis
Light indications
Internal alarms configuration (data base)
Either field or remote supervision – Remote loading
Check of voltages at the equipment own electric power sources
Check and adjustment of analogue measured values
Check of the orders limiter unit
Check of digital inputs and outputs
Check of redundant nodes switching

Remote Maintenance Applications:

REMOTE SUPERVISION: Remote supervision of substations and remote control
equipment. State, values reading, active alarms, internal alarms and Chart Recorder
values collected.
REMOTE LOADING: Remote configuration of control equipment. Tracking and change
of internal operating values and data base full loading.

10.2.2 Functional requirements and added value

Earlier protection schemes may not be considered sufficient anymore and thus also the
increased performance and functionality capabilities can be a driver for change.

Functional integration capabilities make it possible to group all the feeder functions into a
single device thus replacing a collection of elementary elements. This reduces the cost of
the spare parts and copes with the new functional needs. Other drivers for change may be
equipment standardisation

77/80
The new business needs, which require more information, will direct the utilities to
upgrade the existing substations. Therefore information is needed about the industrial as
well as other types of customers, i.e. computerized load forecast, complicated metering
system bulk trading and energy management. The accuracy and the reliability of the data
depend on the utilities/traders. Therefore the data availability gives the utility the chance to
be strong in very competitive field. In next subchapters the major technical issues that
require the upgrading of the existing conventional substation are described.

Examples for need of new functions

In Helsinki Energy (FIN) three most wanted new functions of modern IED’s are: self
checking function, disturbance and event recording and free programmable integrated
functionality. These functions are often also triggers for secondary changes. Self-
checking and disturbance recording functions give possibility to improve reliability and
availability of protection system. Possibility to integrate many functions into same IED
gives financial savings especially in MV level, where separate overcurrent relay,
directional earth fault relay and auto-reclose relay were used before. In HV level modern
relays provide option to combine synchro-check and backup overcurrent relay into
differential and distance IED’s.

10.3 Examples of self-check functions

10.3.1 Hardware

Output relays

The main processor does not have any way of checking if an output relay operated unless
there is a feedback. This feedback may come from voltage and/or current supervision
schemes.

Tripping or closing circuits

The tripping (or closing) circuits work like a series connection in a reliability block diagram
approach. It includes several components that may fail, like battery, wires, relay output
contact, breaker trip (or close) coil, breaker auxiliary contacts (52a and/or 52b), possibly
burden resistors. Some equipment has the capability to monitor the integrity of the
COMPLETE circuit. Therefore this should be called, trip circuit supervision, avoiding the
term self supervision.

78/80
AC circuit

Protections typically have 1 to 12 ac inputs, including current or voltage type inputs. Most
times they come from conventional VTs and CTs. Other times they come from more
modern sensors that provide low level signals (i.e. Low power transformers, Rogowski
coils). In these circumstances they provide analogue low level signals that must be
converted to digital for signal processing. Some protections used spare analogue
multiplexed channels in their internal analogue to digital conversion circuit, to double
check signals sanity by comparing the main voltage reference with another secondary
reference.

Power supply

Power supply is a critical component of an equipment and it suffers more electrical stress
than other low voltage internal circuits. Therefore some ways of voltage monitoring could
be used to detect consumption, voltage, temperature or other parameters that may help to
predict or detect a failure. Obviously it is better prevention than detection.

Batteries
Some equipments use batteries to store information when main auxiliary voltage is not
present. Due to limited battery life, it should be monitored to detect a low voltage before it
actually losses its complete function to store this information.

Communications

Due to the importance of communication in both controllability and monitoring capabilities

that affect product availability, some ways of supervision should be used.
Some supervisions may be done at the physical layer to detect failures such as fibre optic
broken (this may be especially useful in redundant fibre optic schemes).
Other supervisions may be done at upper levels. Some protocols use a “live signal”
periodic message to the master station that may help to detect a ready communication
system that relays not only on physical wires or fibres, but also in the complete IED
communication function. In this sense, modern protocols such as DNP 3.0 and IEC61850
provide naturally schemes for this function. For other simple protocols, such as Modbus,
this function may be implemented in the user application based on simple read/write
simple functions.

10.3.2 Software

In this category it may be include supervisions that affect to software, firmware or

hardware that cooperates to implement the software functions, such as:

79/80
 − Hardware Watchdog.
− Software Watchdog.
− CRC (cyclic redundant check) checking in non volatile memory circuits logs.
− CRC main program checking
− Virtual logic token checking
− Memory utilization
The manufacturer should clearly describe:
− Which component is being tested.
− When and how often is it tested.
− Which message is provided.
− Description of the problem
− Effect and criticality
− What should the user do
− What does the relay do automatically
− Supervision effect on product availability

10.3.3 IEDs response and recommended actions

Major self-test errors also result in the following:

− The critical failure indicator relay is de-energized
− All other output relays are de-energized and are prevented from further operation
− The faceplate Ready LED indicator is turned off
− A “protection out of service” event is recorded
− Internal “critical error” flag set to logic level 1.

Recommended action:
Contact factory for service.
Minor self-test errors result in:
− Faceplate “minor error” LED indicator turned on.
− Internal “minor error” flag set to logic level 1.

Recommended action:
Keep reset key pressed for 5 seconds. Switch auxiliary power off and on. If not cleared,
please contact factory for service.

80/80