Development and Implementation of System Protection Schemes

DEVELOPMENT AND IMPLEMENTATION OF SYSTEM

PROTECTION SCHEMES

Report to Electricity Commission

June 2009

Version 1

Date 15 June 2009

Prepared by David Strong and Michael Green

Client Electricity Commission

Development and Implementation of System Protection Schemes Version: 1

EXECUTIVE SUMMARY

Introduction This report provides an introduction to System Protection Schemes (SPS), provides an outline of some SPS used in Australia, and discusses potential opportunities for the use of SPS in New Zealand to increase transfer limits on parts of the transmission network.

Overview of System Protection Schemes SPS are tools that maintain the power system in a satisfactory operating state following a contingency, and are usually additional to conventional power system control and protection schemes. Secure power system operation means that, following a credible contingent event, the power system remains in a satisfactory operating state. In this satisfactory operating state, all assets remaining in service operate within equipment ratings, system voltages are stable and within allowed tolerances, power flows are stable, and system frequency is within an acceptable tolerance. An SPS may be installed to contain a frequency deviation following a contingency (Frequency Control SPS) or to prevent thermal overloading or voltage collapse following a contingency (Network Control SPS). A Frequency Control SPS (FCSPS) may be implemented to trip generation to limit frequency rise following loss of load, or to interrupt load to limit frequency fall following a large loss of generation. The amount of SPS action is generally a function of the power system conditions at the time and this requires software to calculate the correct SPS action using operational data from a SCADA system. A FCSPS will need to be fast acting, generally within a few hundred milliseconds of the initiating event. It will therefore need to use high speed protection signalling to securely transfer trip signals to the generating units or load blocks. A Network Control SPS (NCSPS) is generally implemented to prevent the thermal overloading of transmission network elements following a specified contingency. The selection of load or generation to be tripped will depend on the location of the transmission element to be protected. If a generating unit is tripped during NCSPS operation, other generation will need to increase output to maintain the network frequency. The increase in output would be expected to be provided mainly by those generating units that have been dispatched to provide Frequency Control Ancillary Services (FCAS). As an example of an NCSPS implementation, Figure E1 shows a transmission corridor with two circuits in parallel. The conventional operational practice would be to limit the flow in the corridor to the rating of one circuit to avoid the overloading of the remaining in service circuit. This is referred to as N-1 as it allows for an outage of one circuit without overloading the remaining in service circuit. If there was generation at A that could be tripped in the event of a circuit outage and generation at or downstream of B that could increase output, an SPS could be implemented to rapidly reduce flow following a circuit outage. With an NCSPS in place the corridor maybe loaded to approximately 95% of the combined rating of both circuits.

A B2

1 Figure E1 Example of NCSPS application The time for an NCSPS to reduce flow in a corridor would generally be in the range of seconds to minutes depending on the pre-event circuit loading.

15 June 2009 Page i


SPS Experience in Australia There are a number of SPS operating in Australia some of which have been in service for many years. These include:

• Basslink (frequency control and network loading);

• Victoria to South Australia interconnector (frequency control);

• Snowy to Victoria interconnector (network loading);

• Tamar Valley Power Station (frequency control);

• Western Sydney (network loading); and

• Snowy to NSW interconnector (network loading).

Runback schemes can also be categorised as a type of SPS. These have been widely used internationally used on generators connections and HVDC links for many years. Examples of these in Australia include:

• Woolnorth Wind Farm; (generation runback for network loading)

• Eyre Peninsula Wind Farms; (generation runback for network loading) and

• Murraylink interconnector (HVDC interconnector runback network loading and voltage control). In Australia, the most complex SPS has been designed to accommodate the Basslink HVDC. This SPS consists of an FCSPS to manage frequency in Tasmania for an outage of the Basslink HVDC, and an NCSPS which enables transmission corridors to be operated up to 95% of the combined thermal rating of each circuit. The Basslink SPS involves equipment installed at two control centres, ten power stations and three major industrial customers. The Basslink SPS enables Tasmania to export up to 630 MW and import up to 480 MW from the mainland. Tasmania has a maximum demand of 1760 MW and a minimum demand of about 900 MW. The Basslink FCSPS has operated successfully on more than twenty occasions since Basslink commissioning in April 2006. There were significant commercial and stakeholder issues that had to be addressed during the development of the Basslink SPS. These issues are discussed in Appendix 2.

Potential Applications of SPS in New Zealand It is noted that Transpower has some existing SPS to prevent circuit overload or transient instability following specific circuit outages. There are also runbacks on the interisland HVDC link which can operate for either north flow or south flow. At Haywards runbacks are initiated by loss of North Island circuits or equipment in the Haywards area, or by loss of communications between Haywards and Benmore. Preliminary analysis has identified further potential opportunities for the implementation of SPS to increase transfer capacity in the New Zealand transmission network. Possible applications for the uses of SPS in New Zealand include:

• the Wairakei Ring;

• the Waitaki Valley;

• Roxburgh to Waitaki Valley; and

• Southern South Island.

15 June 2009 Page ii


CONTENTS

EXECUTIVE SUMMARY I

INTRODUCTION I OVERVIEW OF SYSTEM PROTECTION SCHEMES I SPS EXPERIENCE IN AUSTRALIA II POTENTIAL APPLICATIONS OF SPS IN NEW ZEALAND II

CONTENTS III

1 INTRODUCTION 1

2 OVERVIEW OF SYSTEM PROTECTION SCHEMES 2

2.1 CONCEPTUAL OVERVIEW 2 2.2 SPS DESIGN STUDIES 2 2.2.1 Frequency Control SPS Design 2 2.2.2 Network Control SPS Design 2

3 SPS EXPERIENCE IN AUSTRALIA 5

3.1 SYSTEM PROTECTION SCHEMES 5 3.1.1 Basslink System Protection Scheme 5 3.1.2 Victoria to South Australia Interconnector 9 3.1.3 Snowy to Victoria Interconnector 10 3.1.4 Tamar Valley Generator Contingency Scheme 10 3.1.5 Western Sydney 10 3.1.6 NSW Import Capability from Snowy 11

3.2 RUN BACK SCHEMES 11 3.2.1 Woolnorth Wind Farm 11 3.2.2 Eyre Peninsula Wind Farms 11 3.2.3 Murraylink Runback 11

3.3 NETWORK CONTROL ANCILLARY SERVICES 12

4 APPLICATION OF SPS IN NEW ZEALAND 14

4.1 GENERIC APPLICATIONS 14 4.1.1 Tripping Generation 14 4.1.2 Tripping Load 14

4.2 EXISTING APPLICATIONS 15 4.3 POTENTIAL APPLICATIONS 16 4.3.1 Wairakei Ring 16 4.3.2 South Island 17

5 REVIEW OF SPS SECTIONS OF TRANSPOWER TRANSMISSION CODE 22

5.1 DEFINITION OF AN SPS 22 5.2 SPECIFIC COMMENTS ON CODE 23

6 SUMMARY 26

GLOSSARY 27

ACRONYMS 27 DEFINITIONS 28

REFERENCES 29

APPENDIX 1 AUSTRALIAN INDUSTRY STRUCTURE 30

15 June 2009 Page iii


APPENDIX 2 SPS IMPLEMENTATION PROCESS 33

A2.1 INTRODUCTION 33 A2.2 FUNCTIONAL SPECIFICATION 33 A2.2.1 Reliability 33 A2.2.2 Availability 33 A2.2.3 SCADA System 33 A2.2.4 Software 34 A2.2.5 Operator User Interface 34 A2.2.6 Data Archive 34 A2.2.7 Performance Standard 34

A2.3 TECHNICAL SPECIFICATION 35 A2.3.1 Hardware and Communications 35

A2.4 DUE DILIGENCE 35 A2.5 RISK ANALYSIS 35 A2.6 TESTING 36 A2.7 DOCUMENTATION 37 A2.8 TRAINING 37 A2.9 GO LIVE PROCESS 37 A2.10 POWER SYSTEM OPERATION 38 A2.10.1 Power System Security 38 A2.10.2 Contingency Analysis 38

A2.11 MAINTENANCE AND UPGRADES 38 A2.12 PROJECT MANAGEMENT 38 A2.13 STAKEHOLDER MANAGEMENT AND COMMERCIAL ISSUES 40 A2.13.1 Policy Makers 42 A2.13.2 Industry Representatives 42 A2.13.3 Investors 43 A2.13.4 Retailers 45 A2.13.5 Market Operators and Power System Operators 45 A2.13.6 Suppliers 45 A2.13.7 Commercial Issues 46

APPENDIX 3 NEM REVIEW OF NETWORK SUPPORT AND CONTROL SERVICES 48

15 June 2009 Page iv


1 INTRODUCTION The Electricity Commission of New Zealand engaged David Strong & Associates to provide advice on current practices associated with the development and implementation of System Protection Schemes (SPS1). SPS are tools that maintain the power system in a satisfactory operating state following a contingency, and are usually additional to conventional power system control and protection schemes. Secure power system operation means that, following a credible contingent event, the power system remains in a satisfactory operating state. In this satisfactory operating state, all assets remaining in service operate within equipment ratings, system voltages are stable and within allowed tolerances, power flows are stable, and system frequency is within an acceptable tolerance. This report outlines recent experience gained implementing SPS within the Australian National Electricity Market, particularly in Tasmania, where an extensive SPS was required to accommodate the Basslink HVDC project. A number of aspects of SPS use are discussed, such as the application of technology, resources required, SPS development, and SPS operational experience. Issues surrounding the design, contract negotiations, construction, installation, commissioning and operation of SPS are also discussed. The report builds on previous discussions with the Electricity Commission regarding SPS deployment to support the New Zealand electricity transmission system. To provide more background in the report, Transpower, the New Zealand transmission system asset owner and system operator, was consulted. This was to obtain an understanding of any special circumstances applicable to their situation, determine current practices regarding power system and market operations, and discuss opportunities for increasing the existing power system operating envelope.

1 These schemes may also be referred to as Special Protection Systems

15 June 2009 Page 1


2 OVERVIEW OF SYSTEM PROTECTION SCHEMES Developing an SPS from concept through to commissioning should be approached in the same manner as any other power system project, the difference being that it may involve more parties than a transmission line or substation project. These include offtake customers, generators, and the transmission network asset owner. If there is an independent system operator, it will also need to be involved, as the dispatch process will require SPS operational data. A proposal for an SPS may arise as part of a new generation development, a new major industrial load, or as one of a range of options considered by a transmission planner to address a transmission network constraint. In some cases, an SPS may be the only technical solution.

2.1 Conceptual Overview The first SPS design step is to define the objectives of the scheme, which could be, for example, to:

• contain a frequency deviation following a contingency;

• prevent thermal overloading of transmission circuits following a contingency; or

• prevent voltage collapse. For each objective it will be necessary to define exactly what SPS actions will be necessary following the contingency and how those actions can be implemented.

2.2 SPS Design Studies Power system studies will be required as the first part of the design process, and would be carried out by the SPS proponent or designer. Boundary cases defining the extremes of power system operation should be developed as the starting point for the power system studies. The studies need to determine power system conditions following SPS action to ensure that the power system remains in a satisfactory operating condition following the contingency and SPS operation. This would cover post contingent network voltages, transient stability and frequency. In a weak power system SPS action could have a significant impact on post contingent voltages.

2.2.1 Frequency Control SPS Design A Frequency Control SPS (FCSPS) may be implemented to trip generation to limit frequency rise following loss of load, or to trip load to limit frequency fall following a large loss of generation. In the first instance, power system studies need to be carried out to identify an empirical relationship between the size of the contingency and amount of SPS action required. This will be a function of the power system conditions at the time. At the end of these studies, an algorithm can be developed which is used to calculate the correct SPS action for the power system operating condition at the time. Additionally, the algorithm will also specify the control logic to select the correct generating units or load blocks for tripping, to ensure the power system lands in a satisfactory state following the contingent event.

2.2.2 Network Control SPS Design A Network Control SPS (NCSPS) is generally implemented to prevent the thermal overloading of transmission network elements following a specified contingency. The selection of load or generation to be tripped will depend on the location of the transmission element to be protected.

15 June 2009 Page 2


If a generating unit is tripped during NCSPS operation, other generation will need to increase output to maintain the power system frequency. The increase in output would be expected to be provided mainly by those generating units that have been dispatched to provide Frequency Control Ancillary Services (FCAS). However experience has shown that there is likely to be a response from some of the generating units that have not been dispatched for FCAS. If this is the case and if any remaining generating units are located such that they can “re-load” the transmission asset being protected then this needs to be addressed in the NCSPS design process. The operational characteristics of the generating units which may “re-load” the transmission asset need to be taken into consideration when investigating options to prevent this. Some options are:

• ensure that generating units, which could “re-load” the protected transmission asset, are not dispatched for FCAS raise services;

• ensure that there is sufficient FCAS raise available from generating units on the downstream side of the outage;

• automatically lock the generating unit governors behind the outage following an NCSPS initiated tripping; and

• trip extra generation so that if some “re-loading” occurs the protected transmission asset will still be within its rating.

Parallel power flow paths also need to be taken into account when determining the amount of generation that would need to be tripped. For example, the amount of flow reduction through a protected transmission circuit, from a 1 MW reduction in generation output, will vary due to its location in the network to the protected circuit. The contribution of each generating unit can be determined from load flow studies and then the worst case reduction used in the algorithm. This is referred to as the generator contribution factor. The examples below are used to illustrate these issues. Figure 1 shows a transmission corridor with two circuits in parallel. The conventional operational practice would be to limit the flow in the corridor to the rating of one circuit to avoid the overloading of the remaining in service circuit following a sudden outage of any one of the circuits. This is referred to as N-1 as it allows for one outage. If an SPS is implemented to rapidly reduce flow following a circuit outage the corridor could be loaded to approximately 95% of the combined rating of both circuits. The reason for the 95% is so that the circuit has some thermal capacity available for a short time. If the flow is reduced rapidly (within about ten seconds) the temperature of the conductor will not exceed the design temperature. In example A (Figure 1), for an outage of circuit A-B 1, the post contingent current in the remaining circuit will be slightly greater than the total pre-contingent current due to increased losses. The amount of generation that needs to be tripped at A to prevent overload of the remaining circuit would be the difference between the post contingent circuit loading and the continuous rating of the remaining circuit. In this case the generator contribution factor for the generators at A will be 1.0. The SPS would have a circuit-outage ID for each of the A-B circuits.

A B2

1 Figure 1 Example A – all generation through two parallel circuits Example B (Figure 2) is a different scenario as there is a parallel path and generation in two locations. The NCSPS could be protecting the two D-E circuits and the C-E circuit. For an outage of a D-E circuit the D generators would have a higher generator contribution factor than the C generators.

15 June 2009 Page 3


The D generator contribution factor will be less than 1 as some of the post-contingent output from there will flow through the C-D and C-E circuits. As a consequence a 1MW reduction in the D generator output will reduce D-E circuit flow by less than 1MW. Finally there would be a circuit-outage ID for loss of each D-E and C-E circuit.

ED

C

1

2

Figure 2 Example B – generation sharing between parallel paths Example C (Figure 3) is a network where the flow in a circuit could be in either direction depending on the load and generation pattern. An NCSPS may be used to protect the F-H circuit against an outage of an F-G circuit when the flow in F-H is from F to H, and an outage of a G-H circuit when the flow in F-H is from H to F. In the case of flow from F to H generation at F would be tripped to reduce the loading on circuit F-H. In the case of flow from H to F, generation at H or J would be tripped.

L

F G

H

2

1

1 2

J

L

L Figure 3 Example C – power flow in either direction The circuit-outage IDs discussed above relate to an intact system with all circuits in service. If the system was operating with a circuit out of service, such as during maintenance, the ratio of post to pre contingent circuit loading, and the generator reduction ratios would be different from those of an intact system. The SPS or the specific circuit-outage IDs could be set to disabled for a not intact system. Alternatively there could be a separate circuit-outage ID with the post to pre contingent and generator reduction ratios.

15 June 2009 Page 4


3 SPS EXPERIENCE IN AUSTRALIA This section discusses the Australian experience with SPS and run-back schemes, which are either in service or are presently being developed. The Australian experience has so far covered SPS use to reduce constraints on generation dispatch, meeting customer demand, and maximising transfers across a merchant transmission link and regulated interconnectors. The National Electricity Market which covers the eastern and southern states of Australia has an independent system operator and multiple transmission network service providers. The operation of the NEM is governed by the National Electricity Law, National Electricity Regulations and the National Electricity Rules2, and is centrally managed by NEMMCO. The market dispatch is subject to network security constraints and co-optimisation of energy and frequency control ancillary services (FCAS) markets. Appendix 2 provides an outline of the industry structure and regulatory framework.

3.1 System Protection Schemes

3.1.1 Basslink System Protection Scheme The most complex SPS in Australia is the Basslink3 SPS(1). This scheme comprises a frequency control SPS to manage the Tasmanian frequency for a trip of the Basslink HVDC either under export or import conditions. This SPS also incorporates a network control SPS that monitors 18 AC transmission circuits in Tasmania, ensuring that transmission circuit thermal overloads do not occur in the event of specified outages. Basslink Basslink(2) is a 400kV, monopolar HVDC link with a metallic return comprising 290km of undersea cable, 65km of overhead line with a rated continuous power transfer capacity of 500MW. It has a dynamic rating of 630MW for a period of up to 6 hours a day. Basslink went into commercial service on 29 April 2006. Basslink connects the Australian island State of Tasmania to the State of Victoria. In Tasmania, Basslink comprises a 2km 220kV overhead AC line from Transend4 Network’s George Town substation, to the Tasmanian converter station, and a10km overhead DC line to the coast. In Victoria there is 7km of shore based underground DC cable, 55km of overhead DC line to the converter station and a short section of 500kV overhead AC line to SP-AusNet’s5 Loy Yang substation. Basslink Frequency Controller As Basslink is an asynchronous HVDC link, the Tasmanian system frequency is not synchronised with the main Australian interconnected power system, similar to the North and South Islands of New Zealand. Basslink has been provided with a control action that modulates DC power flows, the objective of which is to align the separate system frequencies in accordance with the different frequency standards that apply in Tasmania and mainland Australia. The frequency controller aims to modulate Basslink power flows so that the relative frequencies lie on the line shown in Figure 4, which illustrates the Basslink frequency controller objective function.

2 http://www.aemc.gov.au 3 http://www.basslink.com.au 4 http://www.transend.com.au 5 http://www.sp-ausnet.com.au

15 June 2009 Page 5

http://www.aemc.gov.au/

http://www.basslink.com.au/

http://www.transend.com.au/

http://www.sp-ausnet.com.au/


46

47

48

49

50

51

52

53

45 46 47 48 49 50 51 52 53 54 55 56

Frequency (Tasmania)

Freq

uenc

y (V

icto

ria)

Figure 4 Basslink frequency controller objective function The key benefits of the frequency control action are that:

• it enables Tasmania to participate not only in the main Australian energy market but also in the frequency control ancillary services markets;

• it improves the quality of Tasmanian frequency control; and

• it supports the Tasmanian Network Control System Protection Scheme (as detailed below). Basslink Frequency Control SPS As the interconnected southern and eastern Australian National Electricity Market (NEM) has a total demand in excess of 30 GW, the connection of Basslink into the Victorian transmission system did not present any significant issues. On the other hand, the Tasmanian power system has a summer minimum demand of approximately 900MW, and a winter maximum demand of 1800 MW. With a maximum capacity of 630 MW, Basslink can represent a significant portion of both the Tasmanian demand (export from Tasmania) and Tasmanian supply (import into Tasmania). Given that Basslink HVDC is monopolar, a pole fault will result in the total loss of DC transfer. If the DC transfer is a significant percentage of the Tasmanian system load at the time, excursions in the Tasmanian system operating frequency would almost certainly breach standards. In order to contain these frequency excursions, a combination of Tasmanian sourced frequency control ancillary services and rapid tripping of generation and/or load is initiated. The SPS is designed to ensure that generation or load is disconnected from the system within 650 milliseconds after the Basslink fault. This scheme is known as the Basslink Frequency Control System Protection Scheme (FCSPS). In the Basslink FCSPS, the amount of generation or load to be interrupted is calculated every 4 seconds and is a function of the Tasmanian demand and Basslink power flow. Basslink Network Control SPS The introduction of Basslink effectively increased the Tasmanian maximum demand by approximately one-third. This was achieved without the need for costly network augmentations, by utilising the “security capacity” of the existing transmission network and enabling operation up to nearly “N capability”. Up to this point conventional practice was to operate the transmission network to its “N-1 capability”. The implication of operating transmission corridors at “N capability” is that the loss of any circuit within a transmission corridor will result in overloading the remaining in service circuits. In the Basslink

15 June 2009 Page 6


example, circuit overloads are removed by the Basslink Network Control System Protection Scheme (NCSPS). The NCSPS consists of a generic algorithm that can be used to prevent overloading of a specified circuit following a specified circuit outage. This generic algorithm is then used with specific parameters to provide a specific circuit outage calculation. There are 40 circuit outages currently implemented in the NCSPS. Most of these circuit outage calculations apply for system normal conditions but some have been developed for specific maintenance or operational outage conditions. There are qualifiers that are used to determine which circuit outage calculations should be running at any time. The qualifiers relate to power system conditions such as specific circuits in of out of service, or direction of circuit power flow. The NCSPS carries out a calculation each four seconds for each of the active circuit outage IDs. This calculation determines what action will be taken in the event of an outage of the specified circuit. Currently the Basslink NCSPS only operates for export from Tasmania, but it also has the capability of providing Basslink import SPS functionality. However the discussion in this report is limited to the export case where a transmission network contingency requires a reduction of generation behind that contingency. The Basslink NCSPS relates to nine transmission corridors and eighteen circuit contingencies as illustrated in the below diagram. If a circuit contingency occurs the NCSPS initiates a reduction in generation behind that contingency sufficient to remove circuit overloads.

Basslink

1

2

3

4

5

6

8

9

7

10

11

12

15

1413

16

18

17

220kV

110kV

Generation

LoadL

L

L

L

L

L

L

Figure 5 Tasmanian transmission circuits relevant to the NCSPS

15 June 2009 Page 7


The loss of generation would normally result in the Tasmanian system frequency falling outside the acceptable range; however, this is prevented through a combination of locally sourced frequency control ancillary services and a reduction in HVDC transfer via the Basslink frequency controller. It is important that the response of the locally sourced frequency control ancillary services does not result in circuit re-loading. This is achieved automatically by inclusion of constraints within the electricity market dispatch software. The amount of generation reduction is calculated every four seconds and is a function of circuit pre-contingent loadings and sets of pre-determined factors that relate:

• generator output to resultant circuit loadings; and

• circuit post-contingent to pre-contingent loading ratios. The algorithms for determining the required NCSPS generation reduction have two important characteristics in that they provide for directional flow in certain transmission corridors and also take account of circuit outages. Extra care was taken when including network topology changes from the normal state, as the complexity of NCSPS logic, algorithms and integration into the constraint formulation for dispatch becomes increasingly complex. A key decision is to determine the scope of SPS operation and under what conditions the power system will operate outside the SPS design boundary; a situation requiring the SPS to be taken out of service. The reduction in generation can be either fast or slow acting depending on the time estimated by the NCSPS for transmission circuit conductors to reach their design operating temperature (and hence have the potential to breach minimum ground clearance requirements). Fast acting control trips the generator fully whilst slow acting control runs back generation by tripping hydro governor solenoids. This ensures that the generators themselves remain connected to the power system. Corridors can be operated up to their “95% N capability” to allow for overloading that occurs between the time a contingency occurs and NCSPS action is completed. This margin was determined based on NCSPS action being completed before the conductors reach the design operating temperature. Terminal equipment ratings were checked against their I2t capability based on fault withstand ratings. Basslink SPS Design Extensive studies across a wide range of Tasmanian power system operational scenarios were conducted to design the FCSPS and NCSPS algorithms. This was also as part of a detailed due diligence exercise to check the power system would remain in a satisfactory operating state following SPS action. The SPS uses a combination of SCADA and protection equipment. The SPS software is integrated with the Transend Network Operational Control System and uses SCADA data to calculate the actions to take in the event of specified contingencies. The tripping signals are issued using protection grade hardware and high speed communications circuits. The FCSPS software sends controls to an RTU at each of Transend’s two control centres to arm generating units or load blocks. In the event of an outage of Basslink a loss-of-link signal is issued by the Basslink control system. This is transferred to the RTUs in the Transend control centres (in Tasmania) by tele-protection signalling units and high speed communications circuits. This loss-of-link signal initiates trip signals to be sent to each armed generating unit or load block. The trip signals are received at the generator or customer trip relays within 60 milliseconds of the signal being issued by the Basslink converter station control system. NCSPS action will be initiated if there is a transmission corridor operating above its firm rating, and the SCADA system receives a circuit breaker open status indication from the RTU at either end of the relevant transmission circuit. This will initiate the issue of trip signals to pre-determined generating units.

15 June 2009 Page 8


The SPS has equipment installed at two Transend Control Centres, ten power stations and three major industrial customers. Acceptance Tests The performance of the Basslink SPS was considered to be essential to maintain power system security. A range of testing was carried out to prove the performance of the SPS. This testing included

• point to point testing of all wiring;

• factory acceptance testing of the software and hardware;

• site acceptance testing of the installed hardware software and communications,

• remote asset commissioning (to prove the performance of the generating unit and load block tripping); and

• live tests to prove the power system response following SPS operation. Operational Experience The FCSPS operated six times during the Basslink commissioning process, three of which were planned SPS tests, and has operated over twenty times since Basslink entered commercial service. The FCSPS has operated correctly each time it has been required to disconnect load or generation from the system. There was an event that resulted in incorrect FCSPS interruption of customer load. This was due to a problem in the RTU used to issue the trip signals. This problem has been addressed and monitoring has been implemented to detect any further hardware problems. There has only been one NCSPS operation and that was caused inadvertently by switching for planned maintenance. Standard operating procedure requires that the NCSPS circuit-outage IDs applicable to a transmission corridor should be disabled prior to switching a circuit in that corridor out for maintenance.

3.1.2 Victoria to South Australia Interconnector The first Victoria to South Australia (VIC-SA) interconnection is via a double circuit 500 kV line from Moorabool (near Geelong) to Heywood near the South Australian border. This 500 kV line also supplies a large smelter load of several hundred MW at Portland (near Heywood). The VIC-SA interconnector had a rating of 500 MW. The loss of both 500 kV circuits could result in a loss of over 1000 MW of load from the Victoria system. A system protection scheme was implemented so that a simultaneous outage of both 500 kV circuits would result in tripping of a 500 MW generating unit at Loy Yang to limit the frequency rise in Victoria.

500 kVSouth Australia

Smelter

Melbourne275 kV

15 June 2009 Page 9


Figure 6 Victoria to South Australia interconnector

3.1.3 Snowy to Victoria Interconnector Victoria relies on high generation import from the Snowy Mountain generation on hot summer days. The transfer capacity from Snowy to Victoria can be increased if required by enabling a network control SPS, installed in the late 1980s. This SPS trips a large industrial load in Victoria to prevent circuit overloading if there is an outage of a 330 kV circuit between Snowy and Melbourne on the Dederang to Murray transmission line. The SPS reduces power flows into Victoria and permits the transmission corridor to operate above its “N-1” limit under normal system conditions. The scheme is only invoked under certain circumstances to prevent pre-emptive load shedding that would otherwise be required to maintain the power system in a secure operating state. The SPS is enabled when the Victorian or South Australian regions of the Australian National Electricity market fall into the condition of Lack of Reserve 2 (LOR2) meaning a situation:

“when NEMMCO considers that the occurrence of the credible contingency event which has the potential for the most significant impact on the power system is likely to require involuntary load shedding. This would generally be the instantaneous loss of the largest generating unit on the power system. Alternatively, it might be the loss of any interconnection under abnormal conditions.”

This scheme is further detailed in NEMMCO’s 2008 Statement of Opportunities (3) under Network Loading Control Ancillary Services, and provides 350 MW of service.

3.1.4 Tamar Valley Generator Contingency Scheme A scheme, referred to as the Tamar Valley Generator Contingency Scheme, is being developed as part of the establishment of a 209 MW Combined Cycle Gas Turbine (CCGT) generator in northern Tasmania. This generator will be owned and operated by AETV Power The Australian Energy Market Commission’s (AEMC) Reliability Panel carried out a review of the Tasmanian frequency operating standards and issued a final report in December 2008 (4). The revised Tasmanian frequency operating standards, along with tightening allowable frequency excursions, specified the maximum effective generation contingency to be 144 MW. The Reliability Panel conducted economic cost benefit analysis in arriving at the revised standards; in particular, the decision included limiting market costs of procuring Frequency Control Ancillary Service (FCAS). This was accomplished by requiring that any generator which could lose more than 144 MW for a single credible contingent event must implement arrangements so that the impact of a contingency would be no greater than 144 MW. In order to connect to the network, the 209 MW CCGT will require the ability to interrupt load for a CCGT trip whenever its output is greater than 144 MW. AETV will implement an SPS and will contract with one or more large industrial customers to provide interruptible load for this scheme.

3.1.5 Western Sydney TransGrid, the transmission network service provider for the State of New South Wales (NSW), proposed in its 31 May 2008 Revenue Proposal (5) to the Australian Energy Regulator, stage four of a 500 kV transmission ring development encircling the Newcastle/Sydney/Wollongong load area. This project is known as the “Western 500 kV Conversion” with a commissioning date in 2009/10. In addition to network augmentation, the project includes network support from embedded generation and contingent load reduction in the Newcastle/Sydney/Wollongong area.

15 June 2009 Page 10


3.1.6 NSW Import Capability from Snowy TransGrid planning studies indicate that transmission constraints will occur on import, from southern NSW, to the major load areas of Sydney, Wollongong and Newcastle following the commissioning of the 640 MW, open-cycle; gas-fired Uranquinty Power Station (located approximately 500 km south west of Sydney and 200 km west of Canberra). To alleviate the constraints, TransGrid have proposed a scheme that would interrupt load in NSW and trip generation south of Yass/Canberra. The generation participating in the scheme can be either within the Snowy Mountains or in Victoria.

3.2 Run Back Schemes Run back schemes can be delineated from SPS in that their action is aimed at alleviating localised issues though reductions in generation or transmission circuit flows where they are controllable. Run back schemes would probably be better characterised as transmission line loading controllers. Australian examples that are outlined below include wind farm generation and a DC interconnector.

3.2.1 Woolnorth Wind Farm The 140 MW wind farm in North Western Tasmania is connected to the Tasmanian power system via a single circuit 110 kV transmission line. While the transmission line rating is determined dynamically; that is, the rating is updated in real time based upon the prevailing weather conditions, there are times when the available wind resource exceeds the transmission line capacity. Under such circumstances, the output of the wind farm is limited to the transmission line dynamic capability through localised generation control.

3.2.2 Eyre Peninsula Wind Farms Unlike the Tasmanian example, there are two separately owned wind farms on the Eyre Peninsula in South Australia that share a common transmission corridor, and at times the combined generation of the wind farms can exceed the single transmission circuit capacity. Under such circumstances, the respective generation from each wind farm output is limited to ensure the transmission circuit thermal rating is not exceeded. Each wind farm output is reduced in relation to its capacity to the combined wind farm capacity. Historically, wind farms in Australia have had the ability to self dispatch. However, as the amount of wind generation and intermittent generation has increased, network security management of the National Electricity Market has become more complex, in particular; forecasting the required amount of centrally controlled dispatchable generation. As a consequence, the Australian National Electricity Market Rules have been recently amended. This is to provide for semi-dispatchable generation under which intermittent generating systems are required to make offers into the market and to follow dispatch instructions.

3.2.3 Murraylink Runback Murraylink was commissioned in 2002 and comprises voltage sourced “HVDC lite” converters and underground cables with a nominal capacity of 200 MW. It interconnects the Victorian 220/110 kV terminal substation at Red Cliffs, to the South Australian Monash 132/66kV substation. Murraylink was developed initially as a market network service provider (merchant link) but has since been converted to a regulated interconnector.



ElectraNet, the State of South Australia’s transmission network service provider states, in its Annual Planning Review (6), that the import capability into South Australia is 200 MW (for system normal summer operating conditions). The capability of Murraylink to inject power into South Australia is influenced by the ability of the Victorian transmission system to supply Murraylink. Under high load conditions it is this factor that limits the amount of power that can be supplied into South Australia by Murraylink. Export capability from South Australia varies considerably as it is dependent on complex interactions between load and generation across electrical regions within South Australia. In its July 2003 submission to the Australian Competition & Consumer Commission on Murraylink’s Application for Conversion to a Prescribed Service (7), the South Australian Electricity Supply Industry Planning Council noted that:

“Unlike other parts of the interconnected network, Murraylink incorporates control schemes that run back and trip Murraylink when incidents occur on either the South Australian or Victorian transmission networks. The Planning Council notes in recent clarification by Murraylink, that VENCorp require Murraylink’s runback scheme to be triggered by outages of any of a number of additional Victorian transmission lines, in order to achieve Murraylink’s claimed transfers.”

The Murraylink run back scheme comprises: • a fast run back with an operating time of 200 milliseconds to 6 seconds that prevents:

• voltage collapse in Victoria; and • thermal overloads in both South Australia and Victoria; and a

• slow runback that results in partial run back after time delay in response to low voltages. As Murraylink flows are controllable, its run back scheme is fairly unique in that it is not associated with interruption of load or tripping of generation. As Murraylink effectively operates in parallel with conventional AC interconnections, its runback simply results in the re-distribution of power flows in the network.

3.3 Network Control Ancillary Services In the Australian context, Network Control Ancillary Services (NCAS) are transmission network services procured by NEMMCO that are critical to the maintenance of secure and reliable operation of the power system. Additionally they also increase the power transfer capability of the transmission network. The Australian National Electricity Rules (8) define NCAS as:

“A service which provides NEMMCO with a capability to control the real or reactive power flow into or out of a transmission network in order to:

a) maintain the transmission network within its current, voltage, or stability limits following a credible contingency event; or

b) enhance the value of spot market trading in conjunction with the central dispatch process.” NCAS are currently procured and deployed by both Transmission Network Service Providers and NEMMCO using the facilities provided by Registered Participants. They are categorised as:

• Network Loading Control Ancillary Service (NLCAS); and

• Reactive Power Ancillary Service (RPAS). An NLCAS rapidly changes active power flow on transmission lines allowing short-term ratings to be utilised. The benefit of NLCAS is that it allows greater utilisation of network capability. Without NLCAS, pre-contingent flows would be limited to ensure more conservative short-term ratings were



not exceeded. Limiting pre-contingent flows can result in a supply shortfall and inappropriate load shedding. NEMMCO procures RPAS under contractual arrangements for ancillary services with appropriate Registered Participants. This is to meet its obligation to maintain power system voltage conditions, so that the power system remains in a satisfactory operating state following a contingency. Without adequate RPAS, the power system would need to be operated more conservatively to avoid voltage collapse. Depending on contingent event location, generation may need to be constrained, or a supply shortfall could arise, resulting in inappropriate load shedding. Currently, NEMMCO is conducting a review into NCAS, or as it is being more broadly termed Network Support and Control Services (NCSC). The objective is to identify and address issues surrounding the current arrangements for the procurement and delivery of NSCS by Transmission Network Service Providers and NEMMCO. This review is further discussed in Appendix 3



4 APPLICATION OF SPS IN NEW ZEALAND The predominant application for SPS in New Zealand appears to be for network control following an outage of a critical circuit. The actions of an SPS would be to trip generation behind a constraint to prevent line overloading or to trip load to prevent line overloading or voltage instability. The issues associated with each of these actions are discussed below.

4.1 Generic Applications

4.1.1 Tripping Generation SPS are well suited where there is a need to operate a transmission corridor at more than its N-1 capacity, and when it is not economic to install additional transmission capacity. The SPS may be feasible due to the intermittent or limited amount of time that the additional transmission capacity is required and other generation is available to meet system peak demand. Wind generation located remote from load centres and the transmission network is an obvious application for SPS or runback schemes. In this scenario it may not be economic to provide the transmission capacity to meet the full rating of the wind generation, given that it is likely to operate at less than its full output most of the time. If an SPS is implemented in relation to a particular network constraint it could potentially allow a higher level of generation behind the constraint than if the transmission network was operated at N-1 capacity. In the case of an SPS that, either trips or runs back generation, there will be a frequency deviation. Other generation in the power system will need to pick up output to bring the frequency back to its target value. It is important that other generating units behind the constraint do not contribute to this increase in output as this has the potential to overload the relevant circuits. The SPS design should include generating unit runback as an alternative to tripping where the time response before transmission circuit integrity is compromised, is in minutes rather than seconds. Hydro generation is particularly suited to this type of runback arrangement as a hydro unit can be easily run down to zero output over the period of a minute. This will have less impact on the power system than opening the generator circuit breaker.

4.1.2 Tripping Load There may be a situation where a large customer plans to locate in a region such that the region peak load cannot be supplied in the event of specific transmission contingencies. In this example the customer may want to consider an SPS to interrupt customer load in the event of a relevant circuit outage. If the customer has a single load block or process, the SPS could be a simple intertrip system where the trip of a relevant circuit directly trips the customer load. If the customer has multiple loads there could be an optimisation algorithm that determines the amount of interrupt for the real time line loading. The first issue will be to identify customers that are willing to have their load tripped. This may be either larger customers that have individual load blocks that can be interrupted, or smaller loads that could be sourced through a demand aggregator. If loads are tripped to prevent circuit overloading or voltage instability, the issue of how soon the load can be restored will need to be addressed. This will depend on the type of load, how soon a customer would want to reconnect its load, and the ability of the power system to supply the reconnected load without circuit overload or voltage instability.



Capacity for the load to be reconnected could be provided by increased generation dispatch on the downstream side of the constraint, or by restoring an outaged line to service.

4.2 Existing Applications It is noted that despite the requirements of Transpower’s recent draft Transmission Code, which appears to restrict SPS use in most situations, Transpower have many existing SPS and runback schemes already in operation. Table 1 lists some of these existing SPS schemes.

Generation Limit(Summer/Winter) Constraint Notes

Te Apiti wind farm

BPE-WDV 110kV 1 or 2

Anchor Generation Karapiro Generation

Karapiro generation is limited to the capability of the remaining circuit plus the Hinuera and Te Awamutu load.

HAM_KPO 1 or 2

Anchor Generation is reduced to zero following a contingency of either the HAM_KPO 1 or 2 circuits

Manapouri Generation

Generation limit of 775 MW Transient stability

Circuit outage intertrips generator at local bus

Maraetai and Waipapa Generation

202/246 MW MTI-WKM 1 or 2

Cobb Generation

COB-STK 66 kV 1 or 2

Coleridge Transient stability

SPS trips all but one of the large units at Coleridge

Table 1 Existing SPS and runback schemes in New Zealand There is also a runback scheme on the HVDC link which can operate for either north flow or south flow and is initiated by loss of North Island circuits or equipment in the Haywards area, or by loss of communications between Haywards and Benmore. This HVDC SPS will runback to a specified MW value depending on the number of interconnecting transformers, condensers, or 220 kV circuits that are planned to be out of service (9) or tripped out of service. It will also optimise the mix of static and dynamic reactive plant to suit the HVDC transfer conditions. This is a very complex SPS and has apparently been operational and trouble free since 1991. It is likely that a similar scheme will have to be developed to accommodate the new HVDC pole when this is proposed to be commissioned in 2012. It is understood that Transpower has been obtaining experience with demand side response through a trial being conducted in the South Island. Meridian Energy also has a runback scheme for wind generation at White Hill in the southern South Island which is connected to a grid exit point 70 km away by a 66 kV line (10).



4.3 Potential Applications Some specific situations where an SPS could be implemented in New Zealand to relieve network constraints are discussed below.

4.3.1 Wairakei Ring The Wairakei Ring (Figure 7) consists of a loop of transmission lines with five transmission connection points on the loop. This loop contains the connection points for existing generation and planned new generation, with the power flow direction variable depending on the generation dispatch pattern. Transpower’s 2009 Annual Planing Report (11) notes that northward power flows could be constrained during peak periods as early as 2010. The Electricity Commission recently approved a Transpower grid upgrade plan involving a new double circuit 220 kV transmission line from Wairakei substation with one circuit terminating at Whakamaru North substation and the other terminating at Whakamaru substation (12). The proposed commissioning date of this project is expected to be mid 2013. The project was approved as an economic investment rather than a reliability investment. This indicates that there is generation to the north of the Wairakei Ring that could be dispatched if generation to the south of the Wairakei Ring was tripped by an SPS following a circuit outage. An SPS could provide an economic option to increase the north transfer capacity in the period prior to the planned new 220 kV circuits. The circuits to be monitored by this SPS would be the 220 kV circuits in the loop and possibly some of the circuits through the Bay of Plenty that form a parallel path to the Ohakuri-Atiamuri circuit. Power system studies would be required to determine which generation would need to be tripped or run back for each specific contingency. The generating units at Wairakei and Rangipo would provide the greatest contribution to northward flow through the Wairakei Ring. These would likely be the preferred generating units to be part of an SPS to increase north transfer capacity. It is noted that there are plans for further generation in this area. This has the potential to create flow constraints on parts of the loop even after the planned upgrade work. A Wairakei Ring SPS could remain in service even after the transmission upgrade to provide increased transfer capacity for further generation south of the ring.



Whakamanu

Ataimura

Ohakuri

Wairakei

Poihipi

Rangipo Figure 7 Wairakei Ring 220kV transmission network

4.3.2 South Island The Transpower Annual Planning Report 2009 considers South Island grid backbone planning issues under the categories discussed below. This discussion is at a high level, based on information contained in the Annual Planning Report 2009, and has not had the benefit of detailed analysis of power system capabilities and generation demand scenarios. Upper South Island and West Coast supply security The Upper South Island and West Coast supply security relates to transmission north of Islington substation to Kikiwa and associated radial extensions. The current Upper South Island demand includes approximately 90 MW in the West Coast region plus 200 MW in the Nelson/Marlborough region. The Upper South Island region appears well served in terms of thermal transmission capacity with 1050/670 MVA (N / N-1), although due to the long distances involved is exposed to potential voltage stability issues. Other than support from local generation and reactive compensation, another option to cater for voltage issues would be to shed load under contingency. Given the present transmission configuration it is likely that options other than load shedding could also provide a net market benefit. Similarly, relatively low cost investments associated with mitigating very low probability, high impact events are likely to provide a net market benefit. North of Waitaki Valley to Christchurch The current load in the Upper South Island is approximately 290 MW, with 820 MW demand in the Canterbury region. The Canterbury region, which goes on to supply the Upper South Island, appears well served in terms of thermal transmission capacity with 2,600/1,800 MVA (N / N-1) from Twizel and Livingstone. However, similar to the Upper South Island region, there is potential for voltage stability issues due to the long transmission distances involved.



Other than support from reactive compensation, another option to cater for voltage issues would be to shed load under contingency. Given the present transmission configuration it is likely that options other than load shedding could also provide a net market benefit. Transmission capacity through the Waitaki Valley The region between Livingstone and Twizel is characterised by minimal load, over 1,500MW of installed hydro generation, and the connection at Benmore of the HVDC link between the South and North Islands. The Annual Planning Report 2009 states that for HVDC northward flow:

“An outage of one of the 220 kV Aviemore - Benmore circuits may cause the remaining Aviemore - Benmore circuit to overload during light and peak loads.

Constraining off Otago – Southland generation and limiting HVDC north power flows will alleviate this issue to some extent.”

The summer/winter capability of each of the Aviemore - Benmore circuits is 202/246 MVA. It may be possible to operate this transmission corridor at up to 190% of 202/246 MVA by implementing a network control system protection scheme, which could trip generation and initiate run-back of northward flow on the HVDC link, should a circuit contingency occur. That is, each line could be operated at up to 95% of its rating and if an outage of one circuit occurred, NCSPS action would reduce the loading of the other circuit to within its rating. This would enable much higher transfers under normal system conditions. This would require either:

• a frequency controller on the HVDC link that provides fast runback of northward flow in response to tripping of South Island generation along with prevention of frequency response of South Island generation that could re-load the Aviemore - Benmore corridor;

• constraining South Island generation that provides response to low frequency to locations that do not re-load the Aviemore – Benmore corridor; or

• the SPS design over tripped generation to take account of any likely re-loading due to short term generator frequency response.

Care would also need to be taken to operate the South Island system in such a way so that loss of load in the South Island does not give rise to overloading of the Aviemore – Benmore corridor resulting from HVDC link frequency control action. The circumstances leading to such an overload would be:

• HVDC Link operating in a northward direction.

• loss of a large South Island load, resulting in:

• the South Island system frequency would tend to rise due to the surplus of generation,

• the HVDC link frequency controller would act to increase transfers from the South Island to the North Island,

• to support this increased transfer would require increased loading of the Aviemore – Benmore corridor.

The Annual Planning Report 2009 states for HVDC southward flow:

“The 220 kV Aviemore - Benmore circuit will overload following an outage of the other Aviemore - Benmore circuit; and

The 220 kV Benmore- Twizel circuit will overload following the loss or outage of either the Aviemore – Waitaki, or Ohau B – Twizel or Ohau C – Twizel circuits.

These issues depend on the amount of power being exported south …….., and are presently managed operationally by constraining on Otago – Southland generation and limiting HVDC south power flows.”



The summer/winter capability of each of the Aviemore - Benmore circuits is 202/246 MVA and the Benmore – Twizel single circuit is 404/493 MVA. Operating the Aviemore - Benmore corridor pre-contingency above 202/246 MVA, that is, possibly operating each of the circuits as high as 95% of 202/246 MVA rather than the conventional 50% (meaning the corridor operates at 190% of its pre-contingency capability), and the Benmore – Twizel single circuit near 95% of 404 MVA/493 MVA may provide an opportunity for a NCSPS. This NCSPS would interrupt load for the loss of a circuit and initiate run-back of HVDC link southward flow. Such a scheme would have the added benefit of preserving hydro storages during low inflow or low water storage situations. Similar implementation issues to those discussed above for northward HVDC flows would apply but an SPS would need to cater for the interruption of load rather than the tripping of generation. Transmission capacity between Roxburgh and Waitaki Valley For high Otago – Southland generation (northward flow) the Annual Planning Report 2009 identifies:

“thermal overload conditions including the outage of one of the:

• two 220 kV Clyde – Cromwell – Twizel circuits, which may cause the remaining circuit to overload; and/or cause the Livingstone – Naseby – Roxburgh circuit to overload; and/or lead to dynamic stability; and/or

• two 220 kV Clyde – Roxburgh circuits, which may cause the remaining circuit to overload.” The Annual Planning Report 2009 also raises the potential for significant wind generation in this area which would likely be connected into Roxburgh. The development of this generation has the potential to exacerbate the amount of constrained off generation in the region. If the amount of wind generation installed to the south of the constraint was such that it became economic to install a new transmission circuit the SPS could then be extended to cover the new circuit. The following analysis assumes that high export from Otago – Southland generation gives rise to either:

• northward flow on the HVDC link that can be reduced or • there is a South Island frequency control service that increases generation without breaching

South Island transmission capability until re-dispatch returns South Island transmission to a secure operating state.

The summer/winter capability of each of the Roxburgh - Clyde circuits is 347/382 MVA and the Roxburgh – Naseby – Livingstone single circuit is 202/246 MVA. Operating the Roxburgh - Clyde corridor above 347/382 MVA (190%) and the Roxburgh – Naseby – Livingstone single circuit near 202/246 MVA (95%) would appear to provide an opportunity to implement a network control system protection scheme. This SPS could operate on the basis that a circuit outage would trip generation south of Roxburgh and initiate an HVDC northward flow run-back. Another option is that the response from an appropriately located South Island frequency control service to increase generation. It is possible that a suitably designed SPS could potentially increase the transfer capacity by over 300 MW. In addition to exploiting wind generation opportunities this SPS could provide another benefit. The SPS could reduce the amount of hydro spill that arises at Manapouri due to any demand reduction at Tiwai and transmission constraints north of Roxburgh, that limit northward transfer capacity. The generation capacity south of and including Roxburgh plus Clyde exceeds 1,500 MW. The peak Otago – Southland demand is 1,100 to 1,200 MW noting that New Zealand peak demands are usually in winter and that the high hydro yield period is spring/summer. These statistics indicate that the operation of the southern New Zealand transmission system and the reasons for constraints is



complex. As a consequence, to develop an SPS covering the transmission circuits north of Roxburgh would require a detailed understanding of the power flows; in particular, flow directions. Transmission capacity south of Roxburgh The region south of Roxburgh is characterised by 750 MW of generation at Manapouri and a peak demand of between 1,100 MW and 1,200 MW. The double circuit 220 kV line from Roxburgh to Invercargill has minimal support from the regional transmission from Roxburgh via Three Mile Hill. The constraint issues raised in the Annual Planning Report 2009 concern meeting regional demand during low hydro yield periods, specifically low availability from Manapouri. An SPS associated with the Roxburgh – Invercargill circuits could permit increased southward flows. Following a transmission line outage, demand would be interrupted either during a circuit outage or until generation from Manapouri could be brought on line. At a high level, the summer/winter capability of each of the Roxburgh – Invercargill circuits is 347/382 MVA, giving an opportunity of increased transfer of about 300 MVA. This assumes that the regional transmission via Three Mile Hill has second order implications and possibly a requirement for voltage support from Manapouri. The advantage of such a scheme would be to preserve high value hydro storage during low inflow periods.



Tiwai

Invergargill

Roxburgh

Manapouri

Clyde

Cromwell

Twizel

Naseby

ThreeMileHill

North Makaewa

Livingstone

Figure 8 Lower South Island 220kV transmission network



5 REVIEW OF SPS SECTIONS OF TRANSPOWER TRANSMISSION CODE This section provides a review of Section 3 and Appendix A of the preliminary issue of the Transpower Transmission Code dated 4 March 2009 (13).

5.1 Definition of an SPS The first point that needs to be clarified is the definition of an SPS. Transpower has broadly defined SPS to include:

• Conventional power system controls that are deployed as a matter of course – these controls support normal operation of the power system and act in isolation from each other in response to local power system conditions that arise from both normal operations and disturbances.

• Defence plans are about minimising the impact of multiple simultaneous contingencies and plans for restoring the system following a major disturbance. Some of the cited tools are normal “defence plan” practice; however, some of the items listed would appear not to be defence plans; for example, power swing blocking is to prevent operation of distance protection schemes and others appear to be more about re-building a power system following a major incident.

Conventional power system controls and defence plans cater for particular circumstances and perform specific roles. They should not be confused with SPS that are designed to respond to a particular contingency event or set of contingency events. The following power system applications are not system protection schemes.

• Fast response excitation and discrete excitation control

• Power system stabilisers

• HVDC control and protection

• Generator dynamic braking

• Static and dynamic reactive compensation

• Automatic switching of reactive compensation

• Controlled line switching

• Tap changer blocking

• System main and backup protection

• Pole slipping protection

• Automatic reclosing

• Power swing blocking

• Tie line tripping The Code could benefit from limiting discussion in the main body to the prime purpose of an SPS being to:

“provide a method of maintaining or enhancing transmission system performance without the requirement for additional reinforcement.”

An alternative definition of an SPS that does not limit their application is:

An SPS is a protection system designed to enhance transmission system capacity without the need for additional reinforcement by protecting power system assets or the power system in



response to a specific contingency or in response to a power system moving outside acceptable parameters.

5.2 Specific Comments on Code Generally the Transpower Transmission Code (Code) provides at a high level a framework for the deployment of Special Protection Schemes (SPS) and a basis for their inclusion in economic analysis of transmission investment options. The intention of the Code is to:

“define the practices, judgements, and standards that Transpower considers reflect GEIP in relation to the matters set out in the Code. These practices, judgements, and standards have been determined with reference to good international practice and consideration and analysis of the New Zealand context. They reflect what Transpower considers would reasonably be expected from a skilled and experienced owner and manager of New Zealand’s transmission grid.”

Further enhancements to the Code should be considered to assist deliver the stated intentions of the Code and assist Transpower meet its Electricity Governance Rules (EGRs) obligation that transmission investment options reflect good electricity industry practice. Section 3.2 of the Code identifies circumstances under which an SPS could be deployed. There would appear to be a mixture of implementation of operational procedures to cater for short term operational requirements and SPS that meet the prime purpose. It should be recognised that proper implementation of an SPS requires very careful planning, analysis, design, liaison, development, testing and commissioning. Even a relatively simple SPS may take at least 9 months to implement with longer times as the complexity and the number of parties involved increases. There should be a distinction between an SPS and an interim operational procedure. Section 3.3 of the Code discusses circumstances under which an SPS will be acceptable. The following requirement would appear contrary to proven practice elsewhere.

“Only schemes that would have a low impact on failure are acceptable. SPSs that would have a high impact on failure are unacceptable”

Following proper process and procedures in accordance with good electricity industry practice is mandatory for the implementation of any SPS not only those that have the potential for wide spread impacts. Experience internationally indicates that complex SPS with potential for wide spread impacts can be implemented successfully and relied upon to perform as designed. The cited exclusions would appear to be too narrow and restrictive – in general an SPS should be in service at all times to avoid unnecessary operator supervision and manual intervention. Section 3.4, Risk assessment, aims to calculate expected probability of failure and the consequences. It is not clear what this will provide and what decisions would arise from the analysis. Part of the implementation process must be to undertake a comprehensive risk – consequence analysis and to have in place strategies for dealing with or at least a clear understanding of the consequences. In particular alternative methods of meeting demand, back-up schemes, defence schemes, and system restoration processes should be established and tested. Section 3.5, Design Principles, generally outlines principles that are sound and can be converted into design standards. The principles include “anticipate the specific power system conditions”. It should be realised that SPS are not forward looking - their operational actions should be based on the prevailing power



system conditions. However, the SPS design should be able to cater for future changes to the power system, for example, transmission augmentations and changes to generation and demand patterns. The meaning of, “but must not use SCADA to detect system conditions that require operation in real time” is unclear and should be further explained if retained in the final Code. This could be associated with using SCADA to deliver trip signals. The Basslink SPS uses Transend’s SCADA system and communications to provide both real time operational data for the algorithms and to receive circuit breaker status signals. However the Basslink loss-of-link signals and load and generation tripping signals are transferred using tele-protection signalling equipment. . It would be advantageous to have a maximum of two high level design principles and articulate a set of design standards so that a proponent of an SPS clearly understands expectations of SPS architecture and hence performance. The principles could be:

• reliability

• robustness against future developments. The following design standards could apply to SPS:

• maximum operating time relevant to the situation;

• designed, maintained, and operated to the same standard as protection systems;

• have duplicate hardware and communications with route diversity to the extent possible;

• have diagnostic and self-check features to detect and raise an alarm and disable action when essential components fail or critical functions or inputs are not operational;

• be totally independent of the operation of any other SPS;

• not interrupt load or generation unless there is an agreement with or regulatory obligation for that customer or generator to be part of the SPS;

• may use SCADA to detect system conditions in order to determine post-contingency actions;

• initiating signals may be derived from either protection or SCADA systems;

• in general there should be duplicate initiating signals; and

• not require frequent (routine) manual intervention (e.g. re-configuration of the SPS). Appendix A.1 states “SPSs can…….. be implemented quickly.” As indicated above, it takes considerable time to implement an SPS; however, they can be implemented in a shorter time frame than transmission line augmentations but probably in the same time as adding a transformer to a substation. Appendix A.1 states “SPSs are not always highly reliable.” SPS must be and can be implemented with very high availability, reliability and predictability of operation. This requirement must be the basic design principle and expressly catered for in every consideration as the consequences of incorrect SPS operation can be wide spread. Appendix A.5.1 deems SPS to be “unsuccessful when they encounter system conditions that they were not designed to protect against (and therefore fail to operate).” This is not a failure of the SPS. This is a failure of the dispatch process. The actions of an SPS form part of the technical envelope of the power system within which the power system operator MUST operate the power system. Operation of the power system outside the technical envelope would most likely be a breach of the System Operator obligations.



Power system operators require assurance that they can rely on the SPS to perform to expectations so that they can meet their regulatory obligations regarding maintaining the power system in a secure operating state. As part of this, power system operators require descriptions of the SPS objective and the permissible power system technical operating envelope. These are an important part of the process so that the power system operator can integrate the actions of the SPS into constraint equations used in the market dispatch algorithm. This ensures that power system conditions outside of the operating design envelope of the SPS should not occur. The relevance of Appendix A.5.2 is not clear other than to try and convey the false impression that SPS are inherently unreliable. It is recognised that SPS can be complex; however, the Code should recognise that complexity does not preclude successful implementation. The Basslink SPS has been successfully implemented in Tasmania. The Basslink NCSPS monitors 18 circuits and covers 40 separate circuit outage cases. The Basslink FCSPS has operated successfully in excess of 30 times. The scheme now operates in the back ground with minimal supervision and is relied upon by asset owner operators and the power system operator for meeting its power system security obligations. The statement that:

“it is a recognised hazard that this degradation can occur without the system operator being aware of the risks being incurred. Maintenance, testing, and extension of complex SPSs can be particularly problematic“

can be seen as an admission of failure of due process and procedures and is not a failure of the SPS. Implementation of any new protection scheme requires coordination with existing schemes and the need to follow established processes. Ideally, SPS should be based upon regions and minimise overlaps of operation. Where there are overlaps then they must be clearly documented and coordinated. Similarly the statement that:

“SPSs are inherently difficult to test, so that even with comprehensive maintenance it is difficult to confirm their overall functionality”

is not without challenge. SPS must go through comprehensive equipment acceptance tests, pre-commissioning and commissioning testing requiring proper coordination and planning. The comment would appear to be only true for so called defence schemes where desk top testing and process exercises are appropriate. There is no reason that a properly designed SPS can not be tested at least as easy as normal power system protection.



6 SUMMARY System Protection Schemes should be considered as an essential tool for transmission companies in maximising network utilisation. New applications for SPS in New Zealand are likely to be for network control rather than frequency control. The constraints on the existing Transpower network and the increasing amount of renewable energy generation with low capacity factors provides potential opportunities to implement SPS as an effective method of increasing transfer capacity in specific areas.



GLOSSARY

Acronyms ACCC Australian Competition & Consumer Commission

AEMC Australian Energy Market Commission

AEMO Australian Energy Market Operator

AER Australian Energy Regulator

AFMA Australian Financial Markets Association

AIG Australian Industry Group

ASX Australian Securities Exchange

BCA Business Council of Australia

BSPS Basslink System Protection Scheme

CB Circuit breaker

CCGT Combined Cycle Gas Turbine

COAG Council of Australian Governments

DNP3 Distributed Network Protocol

ENA Energy Networks Association

EPC Engineer Procure and Construct (contract)

ESAA Energy Supply Association of Australia

EUAA Energy Users Association of Australia

FAT Factory Acceptance Testing

FCAS Frequency Control Ancillary Services

FCSPS Frequency Control System Protection Scheme

MCE Ministerial Council on Energy

MUX Multiplexer

MW Megawatt

NCAS Network Control Ancillary Services

NCSC Network Control Support Scheme

NCSPS Network Control System Protection Scheme

NEMMCO National Electricity Market Management Company Ltd

NGF National Generators Forum

NLCAS Network Loading Control Ancillary Services

OPGW Optical Fibre Ground Wire

RPAS Reactive Power Ancillary Services

RTU Remote Terminal Unit

SAT Site Acceptance Testing

SCADA Supervisory Control and Data Acquisition

SO System Operator



SPS System Protection Scheme

TNSP Transmission Network Service Provider

TPS Tele-Protection Signalling

UFLS Under Frequency Load Shedding

Definitions Basslink An HVDC interconnection between Tasmanian and mainland Australia

operational data Status indications, analogue values and control commands

frequency control ancillary services

Services that are used by the System Operator to maintain power system frequency within the specified limits. FCAS raise services are used to increase the frequency when if falls below the nominal value. FCAS lower services are used to reduce the frequency when it rises above the nominal value.

generator reduction ratio

The amount of reduction in circuit loading provided by a 1 MW decrease in generator output

post-contingency current

The current flowing in a circuit following an outage of another circuit

pre-contingency current

The current flowing in a circuit prior to an outage of another circuit



REFERENCES 1 Strong D, The Basslink System Protection Scheme, The eesa 2007 Residential School in

Electric Power Engineering, February 2007.

2 Green M, Integration of Basslink into the Tasmanian Power System, The esaa 2007 Residential School in Electric Power Engineering, February 2007.

3 NEMMCO, Statement of Opportunities for the National Electricity Market, NEMMCO, October 2008

4 AEMC Reliability Panel, Tasmanian Frequency Operating Standard Review, Final Report, AEMC, 18 December 2008

5 TransGrid, Meeting Customer Needs , TransGrid Revenue Proposal 1 July 2009 – 30 July 2014, TransGrid, 31 May 2008

6 ElectraNet, Annual Planning Review 2008 – 2028, ElectraNet, May 2008

7 Electricity Supply Industry Planning Council, MTC – Application for Conversion to a Prescribed Service, Submission to the ACCC, ESIPC, July 2003

8 Australian Energy Markets Commission, National Electricity Rules

9 Transpower, HVDC: Bipole operating policy, TP.OG 48.02, Issue 4, July 2007

10 Brown R, Wind Generators – Evolving Technology, Presentation by Meridian Energy, April 2008

11 Transpower, Annual Planning Report 2009, Transpower, March 2009, www.transpower.co.nz

12 Electricity Commission, Reasons for Decision set out in Notice of Intention to approve Transpower’s Wairakei Ring Investment Proposal, Electricity Commission, 20 February 2009

13 Transpower Transmission Code, Preliminary Draft, Transpower 4 March 2009, www.transpower.co.nz

14 NEMMCO, Statement of Opportunities for the National Electricity Market, NEMMCO, October 2007

15 System Operations Planning and Performance, Final Determination – Standard for Power System Data Communications, Version No 1.2, NEMMCO, 7 April 2005, www.nemmco.com.au

16 NEMMCO, Network Control Ancillary Service Description (NCAS Description), Version 1.0, 25 February 2008

17 NEMMCO, Review of Network Control & Support Services – Draft determination report, 12 January 2009, http://www.nemmco.com.au/powersystemops/168-0105.pdf

Kema Limited, Technical Standards and Good Industry Practice – Stage 1 – Executive Report, Transpower, 2 October 2008.

Kema Limited, Technical Standards and Good Industry Practice – Stage 1 – Technical References, Transpower, 2 October 2008.


http://www.transpower.co.nz/

http://www.transpower.co.nz/

http://www.nemmco.com.au/

http://www.nemmco.com.au/powersystemops/168-0105.pdf


APPENDIX 1 AUSTRALIAN INDUSTRY STRUCTURE The Australian electricity industry has been restructured over the last twenty years from government owned monopolies to a mix of government and private ownership. There is a National Electricity Market covering the interconnected jurisdictions of Queensland, New South Wales, Australian Capital Territory, Victoria, South Australia and Tasmania. The National Electricity Market Management Company Limited (NEMMCO) was established to be the market and system operator. NEMMCO has overall responsibility for system operation, power system security and market operation. NEMMCO delegates some of its system operator functions to transmission network service providers. As a result of re-organising the overall management and governance of the Australian energy markets, a new entity, the Australian Energy Market Operator (AEMO) has been formed and from 1 July 2009 will take over the functions of NEMMCO and provide for the amalgamation of gas and electricity markets. Each State has a main transmission network service provider that has its own control centre which provides the interface between NEMMCO and participants such as generators, transmission connected customers and other network service providers in that State. A network service provider, either transmission or distribution, will enter a Connection Agreement with any Participant wanting to connect to the transmission network. Within a framework of automatic, minimum and negotiated access standards, the network service providers, in consultation with NEMMCO, are responsible for setting the technical conditions that the Participant has to comply with in relation to the proposed generation or load.

Jurisdictions

COAG MCE

ACCC

AER Jurisdictional Regulators

AEMCCommitteesWorking Groups

Customer organisations•EUA•BCA•AIG

Industry organisations•Grid Australia•ESAA•ENA•ERAA•NGF

Industry participants•Generators•Retailers•Network service providers•Traders

CustomersOther organisations•ASX•AFMA

NEMMCO/AEMO

Figure 9 Electricity industry regulatory framework in Australia



NEMMCO(SCADA)

TransmissionNetwork (RTU) Transmission

Control Centre(SCADA)

GenerationControl Centre

(SCADA)

Power Station(RTU)

Power Station(RTU)

Transmissionconnected customer

(RTU)

Figure 10 Operational data flows in the Australian National Electricity Market Region Maximum

Demand Generation (14)

Queensland 8,677 11,167

New South Wales 14,274 13,322

Snowy 150 3,509

Victoria 10,415 8,363

South Australia 3,331 3,734

Tasmania 1,760 2,521

Table 2 Electricity supply and demand in the Australian National Electricity Market System

operator TNSP Transmission

owner State Comment

VENCorp N Y N Vic Planning

Powerlink Y Y Y Qld

TransGrid Y Y Y NSW

Transend Y Y Y Tas

Electranet Y Y Y SA

Basslink N Y Y Vic-Tas HVDC interconnector

Murraylink N Y Y Vic-SA HVDC interconnector

Directlink N Y Y NSW HVDC interconnector

Energy Australia Y Y NSW transmission and distribution

Country Energy Y Y NSW transmission and distribution

SP-AusNet N Y Y Vic transmission and distribution

Table 3 Transmission Network Service Providers in the NEM



Adelaide

Brisbane

Sydney

Hobart

Melbourne

Victoria

Murraylink

New South Wales

Queensland

SouthAustralia

Directlink

Tasmania

Snowy

Basslink

©Base map data sourced from Geoscience Australia Commonwealth of Australia (Geoscience Australia)

Figure 11 National Electricity Market Interconnectors



APPENDIX 2 SPS IMPLEMENTATION PROCESS

A2.1 Introduction This appendix provides an outline of the process for implementation of a SPS and the issues that need to be addressed.

A2.2 Functional Specification A functional specification will set out the functional requirements for an SPS and should ideally cover

• Algorithms and logic;

• Operating time;

• Reliability;

• Availability;

• SCADA requirements;

• Software;

• User interface;

• Data archive; and

• Performance reporting.

A2.2.1 Reliability High reliability is essential for an SPS as this may be a front line protection scheme, and in some cases it may be expected to operate multiple times per year. The system operator must have confidence that an SPS will operate correctly when required so that the system can be dispatched on the basis of SPS action This generally requires full duplication of tripping circuits and equipment, with route diverse communication to the extent possible. The system should have a high level of monitoring with alarm indication of any failures. This will ensure that the system operator is aware of any failure and can take the appropriate action.

A2.2.2 Availability In most cases an SPS will be in service at all times and will be expected to have a very high availability. The requirement for high availability will influence the design in terms of the reliability of the hardware selected, the software platform, redundancy and spares.

A2.2.3 SCADA System If an SPS is required to calculate actions dynamically then any SCADA system from which it is using operational data must have a high performance and reliability. It is also recommended that it should use a robust communications protocol such as DNP3.



A reliable and robust SCADA system for power system operation must be considered good electricity industry practice for any transmission network system operator. Problems with the SCADA system or RTUs have the potential to significantly impact power system and market operations. The SCADA system should have duplication, with hot or warm standby for rapid failover in the event of a server failure. The SCADA system also needs a function to monitor operational data quality. This function needs to be able to flag bad quality data that fails to update, and any analogues that are considered outside reasonable limits. NEMMCO6 introduced a standard for power system data communications (15) to address the problem of operational data performance. This puts an obligation on all parties responsible for operational data to meet the specified availability and latency requirements. The rules for SPS status during a SCADA system outage will need to be specified.

A2.2.4 Software Software will be required to implement any algorithms and logic. This software is likely to require operational data from a SCADA system as the basis for the calculation of potential actions. The software can either run on a SCADA system or on an Intelligent Electronic Device at either a substation or control centre. The advantage of software running on the SCADA server is that it can be monitored and alarmed to indicate SPS software problems. The SPS will need to run frequently enough to change its potential actions to match the changing power system conditions. A typical SCADA scan cycle is four seconds, and this should be a suitable time for the SPS software to run adequately.

A2.2.5 Operator User Interface It is essential that the control centre operators have good visibility of the SPS. This requires the development of screens that present a clear and concise picture of SPS status. The minimum requirement will be a screen which shows the status of the SPS software and hardware, generating units or load blocks that are selected for SPS action, and status to indicate SPS action.

A2.2.6 Data Archive It is likely that the system operator or owner of the SPS will want to verify that the SPS acted correctly following a power system contingency. Input and output data from each SPS calculation should be archived to allow for analysis of any SPS calculation and operation.

A2.2.7 Performance Standard The performance standard for an SPS will depend on any contractual obligations on the parties involved in the SPS. Where the SPS is owned and operated by one party for the benefit of another, the party receiving the benefit will want some contractual obligations on the SPS operator of the SPS to ensure that it has a suitable level of reliability and availability. The performance requirements may include:

• overall availability

• component availability 6 National Electricity Market Management Company Ltd http://www.nemmco.com.au


http://www.nemmco.com.au/


• time to repair failed component; and

• operational data availability.

A2.3 Technical Specification The preparation of a technical specification may be undertaken by the proponent or a contractor engaged to engineer, procure and construct the SPS (EPC contractor). The technical specification will set out the requirements for hardware and communication to meet the functional specification.

A2.3.1 Hardware and Communications The extent of the SPS hardware will be determined by the functional requirements such as SPS operating time and reliability. The communications will depend on the functional requirements, particularly the time taken to execute a trip signal, from an originating event, to the load or generation to be tripped. The operating time for a frequency control SPS is generally critical and will typically be in the order of a few hundred milliseconds. There is generally more time for a network control SPS, particularly if it is protecting against thermal overload of transmission assets. Operating times for network control SPS can typically be in the order of seconds to minutes. There are various suppliers that can provide signalling units that can send a trip signal between remote locations in a few milliseconds. The timing depends on the signal bandwidth and the number of intervening devices such as multiplexers. Faster communication times can be obtained with higher bandwidth communications. However, generally communication time is likely to be a small percentage of the overall operating time. Fault detection, trip relay and CB operating times will all have a greater influence on the overall SPS operating time.

A2.4 Due Diligence A system operator needs to understand the range of possible power system conditions that will result following SPS operation, and be confident that the power system will remain in a satisfactory operating state after that operation. When the SPS design has been established the network service provider or system operator will generally carry out its own power system studies.

A2.5 Risk Analysis As SPS operation can have power system wide implications an evaluation of the consequences of all possible incorrect operation and coincident system incidents and actions should be undertaken. This analysis not only provides and understanding of the risks associated with the SPS but also gives guidance as to the need for or otherwise and reliance upon other protection schemes or back-up schemes. Such schemes include under frequency load shedding, over frequency generation tripping and transmission circuit over current protection. Some of the types of issues that should be evaluated include:

• failure of SPS action initiation signals;

• failure of load interrupt or generation tripping signals;

• coincident SPS action triggering events; and

• SPS action triggering events coincident with cyclic arming updates.



A2.6 Testing Testing is an essential process to demonstrate the performance of both individual SPS components and overall system. An SPS will involve multiple parties. At a minimum an SPS will involve the transmission network owner and a generator or customer. In some cases it will involve multiple generators and customers. Testing of an SPS may be more complicated than other power system projects due to the number of parties and sites involved. It may not be possible to do a complete end to end test of an SPS as there may be no operation that involves all elements of a scheme. However, the tests must be designed and carried out to prove each individual element of the SPS, and also all the interfaces between elements. The Factory Acceptance Testing (FAT) is normally carried out to prove the hardware and software before equipment is installed. Test cases for the software should be developed to test each possible outcome of each step in a logic diagram. A version of the software may be developed in a spreadsheet and this can be used to check the software developed by the contractor providing the SPS. Any defects in the hardware or software should be identified during FAT and rectified prior to site installation. The assets of any customer or generator providing “interruptibility” services must also be tested. In particular the timing of trip relays and circuit breakers must be tested to ensure that they confirm to the performance requirements. When the hardware and software has been installed at the various sites, Site Acceptance Testing (SAT) is carried out to prove the installation, communications circuits and interfaces between the various elements. SAT can also be used to test the rectification of any issues identified during FAT. The system operator or network service provider may require live tests of the SPS to be carried out to provide confirmation of the SPS itself, or the impacts of SPS operation on the power system. Stage Tests Comment Objectives

1 Initiation signal testing Confirm timing and any logic

2 Component testing Carried out at factory or customer/generator site

3 Factory acceptance testing

Carried out at factory Hardware configuration and performance, software performance

4 Remote asset commissioning

Wiring, trip relay and CB operating time

5 Operational data Validate analogue values, status indications, alarms and sequence of events records

6 Site acceptance testing

End to end test to the extent possible

Prove installation, communications circuits and interfaces

7 Soak testing Impacts on SCADA system operation. Software performance during power system incidents

8 Live tests May be important if size of SPS action is significant in relation to system

Demonstrate power system impacts

Table 4 Test steps



A2.7 Documentation The owner and operator of the SPS should have a full set of SPS documentation, which needs to include the following.

• Functional Specifications;

• Technical Specifications;

• As Built Drawings;

• Hardware instruction manuals;

• Hardware configuration files;

• Software sore code;

• Hardware management software;

• Test Plans;

• Test Reports; and

• Certificates.

A2.8 Training There are a number of stakeholders that will require training in relation to the SPS. These stakeholders include:

• Operators;

• Asset owners; and

• Interruptibility service providers. The training will include:

• Concept and overview of the SPS;

• Operator displays and interaction ;

• Software configuration; and

• Hardware maintenance and configuration.

A2.9 Go Live Process When the testing has been completed and any issues arising from testing that would impact on live operation have been resolved, the SPS may be signed off as ready for live operation. A procedure is also required to move from a test condition, where trip links will be open, to a live condition where all trip links are closed. There should be a procedure or checklist to ensure that all the necessary signed documentation has been received by the relevant parties.



A2.10 Power System Operation

A2.10.1 Power System Security The correct operation of an SPS when a trigger event occurs is essential to maintaining power system security. The system operator must be aware of the status of an SPS when dispatching the system. The most effective way of ensuring that the status of any SPS is taken into account during dispatch is by using the SPS status in a constraint equation. If an SPS is always operationally live, the status will be continually updated in the SO SCADA system. If an SPS is only enabled at specific locations, there is still benefit in having the status indications in the SCADA system so that a failure of any element is alarmed at the time the failure occurs.

A2.10.2 Contingency Analysis The contingency analyser used by the system operator must include any SPS so that any analysis reflects SPS operation.

A2.11 Maintenance and Upgrades It is important that the operator has confidence of the performance of an SPS. In most instances the equipment should have high reliability and maintenance should be minimal. There may be some situations where it is appropriate to carry out routine maintenance or inspect the hardware on a regular basis. If there is full duplication of the equipment one of the two paths can be taken out of service for maintenance while the SPS remains in service with a single path. It may also be necessary to upgrade an SPS from time to time. This may be for an upgrade to the software or hardware, or may be the addition of a new interruptibility service provider. In this instance it may be necessary to have a procedure that allows site acceptance tests to be carried out on a new generating unit or load with the SPS running on the live SCADA system.

A2.12 Project Management Project management of the implementation of a SPS is even more critical than the project management of a transmission line or substation project due to the number of different parties involved. Figure 12 and Figure 13 show a generic flowchart of the process for development and implementation of a system protection scheme. The development of any actual scheme will vary depending on the scope of the SPS, the parties involved, and the implementation time frame.



Software develop

ment

Develop Functional

Specification

Due Diligence

Draft Functional

Specification

Identify interruptibility

providers

Agreement in Principle

Develop Project Plan

Hardware contract

Hardware procure

and construct

Draft project plan and budget

Op Data req

Final Functional Specification

Gen/Cust Contract

Commercial negotiations

Hardware testing

Factory acceptance testing

Power system studies

Tender process

Software contract

Comms Contract

SPS initiation

signal spec

Hardware test plan

Factory Acceptance Test

Plan

FAT report

D

Gen/Cust site works

Comms installation

Gen/Cust installation

testing

Gen/Cust installation test report

A B C

Test SPS signal

Implement SPS signal

PSS/E model

E

SPS concept development

Final project plan and budget

Technical specification

Figure 12 SPS development process – sheet A



D

Installation at central location

Gen/Cust site

installation

Gen/Cust installation

commisisoning

Site Acceptance

Testing

A B C

Soak testing

SAT Plan

SAT Report

Training and

Procedures

Go Live process and

Documentation

Develop Go Live process

Go Live

Issue Resolution

Data archive

and access

E

Gen/Cust assets report

Live tests (if required)

Develop constraint equations

Constraint equations

Figure 13 SPS development process – sheet B

A2.13 Stakeholder Management and Commercial Issues The introduction and deployment of SPS within a power system requires careful setting, meeting and management of all stakeholder expectations. An SPS project involves a coalition of a number of parties with a common objective but a range of their own specific issues and concerns that need to be addressed. Security, reliability, availability, and quality of electricity supply are increasingly crucial to modern economies as they become more reliant on technology and anything that adversely impacts will rapidly erode confidence in management capability.



It is not only technical issues such as generator and consumer access to electricity networks with minimal constraints that need to be taken into account but also policy setting makers’ expectations regarding meeting the challenge of issues as broad as the impacts of global warming and terms and conditions of commercial arrangements between all parties involved in the delivery of a SPS. To ensure successful deployment of SPS stakeholder management at all levels begins at the conceptual stage of development and does not conclude until the SPS have operated successfully. Unsuccessful or partially successful operation of an SPS, either technically or commercially, is not an acceptable outcome with the potential for completely undermining the confidence of all stakeholders in entity management and the ability of the SPS to either meet expectations or efficiently increase the technical operating envelope of the power system. The stakeholders can be categorised broadly under the following groups with Figure 14 showing the stakeholders and their relationships.

• policy makers covering government ministers, departments, and regulators;

• industry representative and lobby organisations;

• investment decision makers and asset operators covering generators, large consumers, and transmission network owners;

• retailers;

• market and power system operators and managers; and

• suppliers.

Policy Makers

Regulators

GovernmentDepartments

SYSTEM PROTECTION SCHEME STAKEHOLDERS

Investors

Generators

TransmissionOwners

Consumers

Market & Power SystemOperators & Managers

InsurersFinanciersRetailers

IndustryRepresentatives

Suppliers

Policy Makers

Regulators


Policy Makers

Regulators


SYSTEM PROTECTION SCHEME STAKEHOLDERS

Investors

Generators

TransmissionOwners

Consumers

Investors

Generators

TransmissionOwners

Consumers

Market & Power SystemOperators & Managers

InsurersFinanciersRetailers

IndustryRepresentatives

Suppliers

Figure 14 SPS Stakeholders and relationships The list of stakeholders is quite broad and each will need to be managed at different stages of the introduction of SPS. The party responsible for managing individual stakeholders and at when that takes place should be established at an early stage. It should be recognised that stakeholders can have a dual role of being also stakeholder manager. In fact, the stakeholder and manager roles can be reversed as the development process progresses.



A2.13.1 Policy Makers Policy makers’ interests lie in the areas of ensuring that the appropriate incentives and processes are in place for the delivery of goods and services at a quality and price that furthers broader community economic development. Establishing an environment in which public and industry perceptions are that policy settings are appropriate is essential to successful implementation of changed approaches to delivery of services; particularly in situations where a cost – service level optimisation has been implemented. Governance of the decision making process needs to be well understood, clearly allocating responsibilities and accountabilities, and delineating between rule setting and rule compliance oversight. The key elements for policy makers are essentially:

• How does it support the policy settings and objectives?

• How will it improve or be different from the current situation?

• How does it enhance the environment for investment?

• What are the impacts on the wholesale market and trading?

• What introduced risks are there and how are they managed?

• Why do we need to take this approach?

• What are the benefits and who are the beneficiaries?

• What are the costs (financial and service levels) and who bears them?

• How are the costs recovered? The key to success is to establish the processes for communicating and developing relationships with policy makers and their advisers. The communication issues that should be considered include:

• What are the forums to liaise?

• Who is involved in the forums?

• What are the levels of communication?

• By what methods are messages delivered and any issues or concerns allayed?

• Who will deliver the messages? By their nature, managing and operating electricity markets and power systems is highly technical with its own unique concepts, language and jargon. As a result it is very easy to convey mixed messages to policy makers and to instil an atmosphere of uncertainty and moves toward safety and retention of the status quo. Conveying highly technical based information in a non-technical environment in a readily understood and digestible format without introducing misconceptions requires careful consideration and the ability to have open and two way communications.

A2.13.2 Industry Representatives Industry representatives to a large extent have similar if not the same issues as policy makers but with emphasis on the particular sectors that they represent. Experience indicates that representative concerns are largely around impacts on costs of service provision, from whom costs are recovered and quantity and quality of service delivered. Industry representatives have strong linkages back to policy makers hence common forums and methods of message delivery and liaison can be beneficial.



A2.13.3 Investors The category of investors includes generators, large industrial consumers and transmission asset owners – Transpower. It is within this grouping that SPS implementation stage efforts are concentrated following the establishment of the policy settings and operational rules. Within the investor category technical and commercial issues are resolved and contractual relationships are established. Generators seeking access to the power system make locational decisions based on a number of inputs covering energy source availability and cost, cost of technology to utilise the energy source, cost and availability of supporting services, and the level and cost of access to the power system. Transmission connection and access issues are not generally the main determinant for generation locational decisions. SPS do provide the opportunity to offer a lower cost of connection at an acceptable level of access to the power system commensurate with any hedging arrangements with retailers or consumers. Similar arguments applicable to generators also apply to new entrant large industrial consumers and existing consumers seeking increments to their demand. The differences being:

• exposure to spot markets can, in addition to management through financial or physical delivery contracts with generators, be managed through load demand participation and profiling in the wholesale market. This participation can be manual in response to pricing signals or automated in response to power system access constraints or contingency events.

• consumer value of unsupplied energy can significantly exceed the prevailing spot price and hence consumers have strong incentives for the duration of the demand response to be minimised and assurances of maximum supply restoration times

Generators and customers can both be either beneficiaries of an SPS or participate in delivery of the services required to support the SPS. Benefits accrue through power system access arrangements and participation is through provision of trippable or run-back generation and interruptible loads. The benefits of an SPS can accrue to either an individual market participant or to a coalition of parties connected to the power system. The easier it is to ascribe benefits to a single entity the easier it is to implement an SPS and for the associated parties to enter into agreed commercial arrangements. Multiple agreements between numbers of parties are a normal consequence of situations where there is a single beneficiary and a number of participants involved in the delivery of the service. In situations where there are multiple beneficiaries and multiple participants to service provision the complexity of contractual arrangements becomes such that a regulated obligation may be necessary if a representative of the coalition of beneficiaries to act on the behalf of the beneficiaries cannot be agreed. Normally, the entity in the best position to act as the proponent of the SPS would be the transmission asset owner. In offering to generators and consumers connection to the power system at agreed levels of access transmission asset owners have the option to utilise the “security capacity” of their transmission system through SPS applications thus increase transfer limits above those applied conventionally. In this context “security capacity” refers to that capacity of the transmission system that is traditionally set aside to allow for continued transport of electricity following the loss of one (or even more in some locations where consumers ascribe a very high value to energy otherwise not consumed) of a number of parallel transmission circuits. Foremost, electricity transmission asset owners have to be certain that they can comply with their regulatory obligations, licence conditions, and connection agreement provisions as applicable to service delivery and access reliability standards. The regulatory regime and connection agreements have to be such as to provide transmission licence holders the opportunity to meet those obligations. From an operational perspective the capability of the transmission system in conjunction with possible support services and processes must provide transmission asset owners the reasonable opportunity to plan with certainty and perform required asset maintenance, replacements, and augmentations.



In introducing SPS into their toolkit for planning and developing the transmission system, transmission investors also require certainty as to recovery of all costs of implementing such schemes inclusive of the appropriate rewards for the level of risk assumed. Operational personnel must have confidence in SPS efficacy so that operationally SPS become in the “back ground” and do not require special oversight and supervision. Any scheme that requires micro management by operational personnel only serves to act as a distraction to efficient operation and appropriate responses to power system incidents. In this context key stakeholder groups within the transmission asset owner are the asset managers and operators. Their support of and confidence in any SPS are essential to successful implementation. Experience indicates that there could be strong scepticism regarding the efficacy and reliability and the operational complexity of schemes. Their involvement in the design, implementation, training, commissioning, and ongoing maintenance and upgrades proved to be effective in gaining their support for SPS. Figure 15 below is a poster providing an example of the scepticism about the Basslink SPS, but following successful implementation and operational experience there has been substantial praise and acknowledgement of success.

Figure 15 Scepticism of SPS Success The board of directors of a transmission company understand the risks and rewards of owning and operating the assets of the company; however, the introduction of an SPS potentially increases the risk profile of the company. This increased risk exposure rests with the potential wider power system consequences of incorrect operation of an SPS. As a consequence, directors will require a clear understanding of the operation of the SPS and the related commercial, regulatory, and contractual relationships. In addition, directors will require assurance that their fiduciary duty of care to the shareholders is fulfilled. The board will require demonstration that all reasonable and possibly a higher standard of best efforts have been undertaken by the company to prevent foreseeable, incorrect operation of an SPS. This places the company in a position to mount a successful defence against any potential negligence action. The board will also require that reasonable avenues for limiting liability have been explored and acted upon and that appropriate insurance coverage has been secured. Limits to liability would be normal provisions of connection and access agreements and SPS related agreements with the beneficiaries and participants. However, unlimited liability exposure could remain to third parties that do not have a contractual relationship. In these circumstances a deed of indemnity from the beneficiary is an option for consideration. Company insurers require full disclosure of asset related risks within the insurance proposal. Similarly to director’s requirements, the insurance proposal is likely to require SPS details covering purpose, actions, design principles and operation and maintenance practices.



A2.13.4 Retailers Retailer interests lie in managing their risks around contracts with consumers. Those risks involve both wholesale price exposures and the nature of the product offered to the end consumers. SPS are not about any shortfalls in wholesale electricity productive capability but the capability of the integrated “transport” system. Thus retailers are concerned about the opportunities that the SPS presents for bundling together groups of consumers to provide differentiated services. The bundled impact on retailer’s wholesale electricity and transport costs for offering differentiated services and a clear description of the services that the retailer and consumers can readily understand is essential. The retailer will need to understand what services are available, what opportunities are there for entering into specific arrangements, what certainty does the SPS deliver, and with whom do they enter into such arrangements. In an environment of competing retailers within the one geographic region the provision of different service levels can represent specific challenges associated with potential “free rider” issues. The future development of so called “smart grids” and opportunities presented by sophisticated two way communications and customer information systems could eventually provide opportunities for better delivery of differentiated services.

A2.13.5 Market Operators and Power System Operators Market operator and power system operator interests concern the impacts on market outcomes and the ability to maintain the power system in a secure operating state. Market operators, whilst mainly concerned with spot price outcomes being in accordance with rules, will want to understand how the SPS could influence the outcomes of the spot market and whether there are opportunities for unusual outcomes or abnormal fluctuations. Power system operators require assurance that they can rely on the SPS to perform to expectations so that they can meet their regulatory obligations regarding maintaining the power system in a secure operating state. As part of this, power system operators require descriptions of the SPS objective and the permissible technical operating envelope. They also require demonstration that the response of the power system to the initiating contingency and subsequent SPS action across the range of power system dispatches within the technical operating envelope is such that the power system is stable and resumes operation in a satisfactory operating state. These are important parts of the process so that the power system operator can integrate the objectives of the SPS into constraint equations used in the market dispatch algorithm. This ensures that power system conditions outside of the operating design envelope of the SPS should not occur. Occurrence otherwise would mean that the power system operator has not met its power system security retention obligations.

A2.13.6 Suppliers In this context suppliers refers to the party or parties that engineer, procure, and construct (EPC) the SPS. The SPS owner could engage either a single EPC contractor or project manage a number of contractors with specific expertise. This section is not about the advantages and disadvantages of different procurement options but outlines issues that can be encountered during implementation phase. In general an SPS will require assets to be installed within the premises of a number of SPS participants. In addition for the need to coordinate SPS installation and outage requirements with the normal production operations of participants, issues for resolution include EPC access to the site,



professional indemnity insurance, public liability insurance, workers compensation, site supervision by the participant and participant insurance coverage. The SPS will require interfacing with participant assets and for insurance reasons, the participant may require that it undertake the associated works. The contractual relationships must clearly allocate responsibilities and liabilities and coordinate installation, testing, commissioning and going live processes. The owner of an SPS will want to be I n a position to upgrade software functionality and write their own code requiring access to the source code. If a proprietary software is used the supplier may want to protect its intellectual property and not be prepared to provide the source code other than to have it held in escrow on behalf of the SPS owner. This can expose the SPS owner to commercial disadvantage under circumstances where the original supplier is the only option to implement changes.

A2.13.7 Commercial Issues Several commercial issues have been covered previously with this section addressing the remainder. The previous discussions have to some extent been based on ownership and responsibility for the SPS resting with the regulated transmission service provider but this may not be the case. A key issue is to establish SPS asset ownership and responsibility boundaries. There are a number of options of both ownership and boundaries. Potential owners include the transmission asset owner, the power system operator, the proponent or beneficiary of the SPS, and a service provider. SPS asset boundaries include the data acquisition system, telecommunications, central assets and software, and participant site assets. Additional assets and boundaries can arise from requirements to provide SPS status information back to the power system operator. Ownership and responsibility boundaries are subject to negotiation but usually are determined by the party in the best position to operate and maintain the assets and manage the associated risks. Other ownership issues include:

• are enhancement of competition;

• barriers to access to participate in an SPS;

• avoidance of SPS ownership proliferation;

• avoidance of dilution of allocation of roles and responsibilities associated with delivery of power system capability and management of operational security; and

• minimising the number of required contractual relationships. To enable the SPS owner to trip or interrupt participant generators or loads requires the parties to enter into agreements to do so. These participation agreements provide for performance obligations, equipment inspection and testing requirements, reporting requirements, service availability expectations, participation withdrawal protocols, liability allocations and caps, restoration expectations, fees or compensation, load block or generator characteristics covering maximum and minimum amounts available for participation, and withdrawal durations following a response SPS action. In addition, these agreements would cover arrangements for assets not owned by the participant to be located within the premises of the participant. For each time that participant services are called upon a service delivery certificate could be required detailing performance issues such as:

• date and time of performance;

• time to order to resume consumption or generation;

• load or generation before service;

• load or generation immediately after service;

• load or generation profile to full restoration;

• energy not served or delivered; and



• assessment of performance. The SPS proponent may have to procure participant services. They may approach potential participants directly or embark on a tender process for the provision of participant services. The procurement process can be lengthy and not well understood by potential participants for a number of reasons including being a new experience, uncertain value or cost of participation, impacts on insurances, likelihood of performance, unknown timing of performance, priority order in providing services, requirement for confidentiality, and supply restoration or reconnection times and procedures. It is needless to say that the time required to procure these services and enter into participation agreements should not be under estimated. A possibility is that once an SPS has been established that a prospective participant may approach the SPS owner to have their facilities integrated into the SPS. This eventuality should be provided for in the design. Integration of a new participant will incur costs to cover due diligence SPS and power system performance studies, negotiation of agreements, integration into the software, and supporting equipment covering central, communications, and participant site assets. The SPS owner will incur a variety of costs that if they are the beneficiary will be covered by the benefits. If the SPS is provided as a service, then the cost will need to be identified and recovered fro the participants and beneficiaries. These costs include:

• payments to SPS participants possible in the form of availability, performance and loss of market trading income;

• allocation of corporate costs possibly based on a published cost allocation mechanism;

• recovery of SPS related additional corporate costs;

• asset operation, maintenance, supervision and renewals;

• regular testing and inspection;

• software licence fees, maintenance & supervision;

• preparation of service delivery certificates;

• post SPS action evaluation of:

• participant performance,

• SPS actions, and

• power system response;

• participant agreement supervision;

• performance improvement initiatives; and

• evaluation of power system technical operating envelope.



APPENDIX 3 NEM REVIEW OF NETWORK SUPPORT AND CONTROL SERVICES Network Control Ancillary Services are transmission network services procured by NEMMCO that are critical to the maintenance of secure and reliable operation of the power system, by providing the capability to:

• Maintain the transmission network within its current, voltage or stability limits following a credible contingency event; and

• Enhance the net value of spot market trading by increasing transmission network power transfer capability within those limits.

In its NCAS description (16), NEMMCO describes network control ancillary services as:

• Network Loading Control Ancillary Service (NLCAS) used to change the active power flow on Transmission Power Lines rapidly allowing short-term ratings to be used, and

• Reactive Power Ancillary Service (RPAS) to maintain the power system’s voltage condition, so that the power system remains in a satisfactory operating state. Without adequate RPAS, the power system would need to be operated more conservatively with constrained generation or inappropriate load shedding.

NEMMCO is conducting a consultation to inform its review of the now termed Network Support and Control Services (NSCS) which have a broader scope than NCAS. Historically, both transmission network service providers and NEMMCO procure NSCS under various legislative instruments and obligations and the outcomes they seek to achieve, are in many ways difficult to distinguish. The objectives of the review are to:

• identify and address issues surrounding the current arrangements for the procurement and delivery of NSCS by transmission network service providers and NEMMCO; and

• evaluate and make recommendations on potential alternative arrangements for the more efficient procurement and delivery of NSCS.

Under the proposed NSCS definition, services that might qualify as maximising net economic benefit by increasing secure power transfer capability include:

• Automatic generator run-back schemes;

• Automatic load shedding schemes (such as NLCAS); and

• Generators with automatic generation control (AGC) enabled in tie-line bias mode that compensate for the output variations of intermittent (wind) generation located in their area, to allow the tie-line with other areas to operate up to a higher limit using a less conservative margin (although such a scheme might also create local frequency control ancillary service requirements and increase costs.

It would appear that the review is leaning toward allocating the responsibility for procuring NSCS to transmission network service providers (17). All consultation details and relevant documents may be found on http://www.nemmco.com.au/powersystemops/168-0089.html


http://www.nemmco.com.au/powersystemops/168-0089.html

Development and Implementation of System Protection Schemes

Documents