Child pages
  • Power Engineering Society + Power Electronics Society

Coordination of Protection and Automation for Future Networks

CIGRE B5.43

 

 

 

 

Version 11a Décember,6  2012

 

After WebEx WG meeting Oct, 26, 2012

 

Incorporating contribution received before Oct,22, 2012 and modifications incorporated during the October WebEx meeting


 

CIGRE B5.43

1.Preface

2.SCOPE

3.Abbreviations

4.Drivers for Change

5.Vision of the Future Grid

5.1Bulk Transfer and Super Grids

5.1.1Definition

5.1.2Example of South African supergrid– Adam

5.1.3Links and DC Networks - Volker

5.1.4Multi-terminal DC lines and DC busses – Volker

5.2Regional sub-transmission grid  - Adam

5.2.1Definition

5.2.2Subtransmission aspects in future networks - Adam

5.3Distribution grid – Mika

5.3.1Network Structure  – Volker

5.4MicroGrids – Mark, Tianshu

5.5Local Energy Networks  - Paul Mydra

5.5.1Design Criteria

5.5.2Components of Local Energy Networks

6.Connection of Various Energy Sources

6.1.1Connection of Distributed Energy Resources to Distribution

6.1.2Implementation of Tap Changers in Distribution – Mika

6.1.3Connection of various Energy Resources to the transmission network

6.1.4Network Dynamic Characteristics

6.1.5Effect of response of inverter-based systems – Paul Myrda

7.Evolutions in AC Systems

7.1Network Topology and Substation topology (Network Structure) to be assigned

7.1.1Substation Topology

7.2Increment of power flow in a transmission corridor

7.3Six-phase systems – Mark / Jorge Cardenas / Alex

7.3.1Six Phase Fault Analysis

7.3.2Six Phase Symmetrical Components

7.3.3Fault Combinations

7.3.4Challenges in the application of protective relays on a six phase system

7.4FACTS – Sankara

7.5Stability

7.6– how much extra weight / stress can existing towers support?

7.7Interface requirements for Generators

7.8Overload Management of lines - Jorge

7.8.1Context

7.8.2Sag Check

7.8.3Calculation of the Ampacity {6.4[1]}

7.8.4Impact of a LOMS {6.4[1]}

7.8.5Brazilian Viewpoint {6.4[2]}

7.9Networks with increased Cable Lengths

7.9.1Distribution grids– Mika (contributions from Lars)

7.9.2Transmission networks – Volker

7.9.3Automation in distribution networks - Mika

8.Impact of Technology Trends on Protection and Automation of Primary and Secondary systems

8.1Impact of HVDC on AC protection - Volker

8.2Impact on network and substation automation - Volker

8.3Protection of HVDC networks - Volker

8.4Protection of local LV DC networks

8.5Impact of FACTS on protection and automation – Sankara

8.5.1Reach of Distance relay

8.5.2Effect of Load Angle

8.5.3Relay Operating time

8.5.4Unsymmetrical Faults

8.5.5Symmetrical Faults

8.6Effect of Power Electronic Conversion on Protection

8.6.1Transmission and Sub-transmission – Nabil + Volker

8.6.2Distribution – Nabil

8.6.3Coordination problems

8.6.4Solid State Transformer – Sankara

8.7Super Conducting networks / A3.16 –Sanjay & Teruo (in process)

8.8Fault Current Limiters – Paul Mydra

8.8.1Background

8.8.2Reasons for Increased Fault Currents

8.8.3Impact of FCL on protection - Johann

8.9Evolution of Communication Technology

8.9.1Communication requirements for coordinated protection and automation functions

8.9.2Wide Area Communications - Adam

8.9.3Cloud Computing

9.Developments in Protection and AutomationTechnology

9.1Evolution of SAS Architectures  - Transmission – Mark to provide drawing and words  - Distribution - Mika  - move bullet points from end of 8.5.5

9.2Coordination between Primary and Backup protection – Andrei / Jorge Cardenas / John Cuifo

9.2.1Distribution / Sub-Transmission (<69kV) [Ciufo]

9.3Use of IEC 61850 and future extensions  - Emiliano

9.3.1Future system engineering tools  - Valeriy

9.4Adaptive Protections

9.4.1Definition of Adaptive Protection

9.4.2Possible applications of Adaptive Protection in Future Networks

9.4.3Issues of Adaptive Protection Application in Future Networks

9.5Adaptive tripping on phase-to-phase faults – Alex

9.6Advanced Adaptive reclosing

9.6.1Adaptive single-pole autoreclosure

9.6.2High-frequency probation

9.7Sensing technology – T.W. Cease

9.8Application of SIPSUse of synchrophasor measurements - Mark - Include application of Synchrophasors for Distribution State Measurement - Synchrophasor-based Linear State Estimation

9.9Faster than real-time contingency analysis – Paul Myrda - Marek

9.10Earthing Policy changes – Jorge Cardenas

9.10.1Introduction

9.10.2Ungrounded Systems

9.10.3Solid Grounding

9.10.4Resistance grounding

9.10.5Earthing Policy during islanding

9.11Implications for Testing - Alex

9.12Future redundancy

10.Future Network Use Cases

10.1System instability – BR

10.2Large loss of generation – underfrequency – (BR)

10.3Transient stability – (BR)

10.4Transient stability / angle instability / islanding – Mota / Adam / Alex

10.5Voltage Stability – Peter

10.6Causes of cascading power system events [1]

10.6.1Prevention against cascading power system events

10.7Future Distribution System

10.7.1Grid-connected

10.8Island grid – all inverter based generation – Peter

10.9Long duration faults (ride-through & VAR support issues) – Nabil

10.10Automation based on Asset condition – Peter/Alex (switching to protect assets)

10.11Adaptive protection responding to network changes resulting from automation – Volker / Andrei / Tianshu

10.11.1Case : Skegness-Boston DLR Scheme [NE of England]

10.11.2Case : Adative distance protection using network topology

11.Transition to the Future Grid

12.Bibliography

13.Appendix 1 – Définitions

14.Appendix 2  Definitions of Wind Turbine Generator Types

15.Appendix 3 – Low Voltage Ride Through Capability


1.     Preface
 

The structure of the electric power grid is set to change dramatically over the coming years with the implementation of distributed generation and new technologies to improve efficiency and capacity. These will pose many challenges and opportunities. This report looks at the likely model of the future grid and its prospective components, as well as identifying some of the requirements for protection and automation.

2.     SCOPE

This report identifies and discusses explains issues for protection and automation related to Future Networks and then discusses envisioned solutions for these issues.  The picture of the future of the grid is beginning to take shape.  The utility industry needs to be proactive and take steps to shape the network to best meet its needs.  To accomplish this, the engineering community needs to identify what functions are coming to the grid and have the right tools and technologies available to best integrate everything together. This is easier said than done, and there will be challenges along the way to the network of the future, as several of the technologies may have only been used in trials so far. Whilst the principles and building blocks, such as IEC 61850 for communications are in place, the infrastructure for protection and control systems, particularly in lower voltage system levels may not be and will require investment.

New changes are required also to provide the necessary education and knowledge for protection and automation engineers to cope with these technological changes. Specifically, new skills are demanded related to network and software engineering, as well as management proficiency and tools, in addition to traditional electrical and electronics engineering.

3.     Abbreviations

 

DER Distributed Energy Resources

DG Distributed Generation[MGA1]

RES[MGA2]Renewable Energy Sources – included in DER

 

DSAS Digital Substation Automation System

 

FACTSFlexible AC Transmission System

 

FRTFault Ride Through

 

GTOGate Turn On

 

HVDCHigh-Voltage DC

 

IGBTIsolated Gate Bipolar Transistor

LCCLine Commutated Converter

MTDCMulti-terminal DC

NCITNon-Conventional Instrument Transformer

PVPhoto Voltaic

RAS Remedial Action Schemes

RESRenewable Energy Sources

 

SIPS System Integrity Protection Schemes

SPS Special Protection Scheme

SVCStatic Var Compensator

SSSC Static Series Synchronous Converter

STATCOM Static Synchronous Compensator

 

TCSCThyristor Controlled Series Capacitor

 

VSC Voltage Source Converter

WAPS Wide Area Protection Schemes

WTGWind Turbine Generators

4.     Drivers for Change

The goal of making use of alternative energy resources is resulting in the restructuring of the electric power grid as we know it today.  It is estimated that, over the next 20 years, millions of windfarms and multi megawatt photovoltaic plants will be connected to the transmission network and that tens of millions of distributed energy resources will be connected worldwide to the grid – both to transmission and distribution networks.  Existing storage technologies, like pumped storage schemes, will be complemented by new technologies such as batteries and flywheels. Distributed Energy Resources (DER) will be connected mostly at the lower network voltages, many of which were not designed to cater for embedded generation and pose challenges for conventional overcurrent-based protection schemes[VL3]. This will create stability and power flow reversal issues to be resolved. Protection and control at all voltage levels, both locally and over wide areas must be co-ordinated and new protection and automation applications are likely to emerge. With these additions and the changing face of generation, transmission and distribution, the network of the future poses a number of challenges and opportunities.

 

Generation trends will likely change from predominantly large base load power plants stations connected at the transmission level to a mix of large dispatchable stations plants at transmission level along with a large amount of embedded, generation with varying degrees of dispatchability, much of it from RES, and maybe some power storage. These embedded RES provide opportunities, or problems, depending upon one's viewpoint such as islanding of parts of the network under certain conditions. However, some DER do not possess the fault ride through capabilities of conventional generators traditional generation sources and may compound stability issues during large fault disturbances and faults. Despite the limited dispatchability today of many RES, the main control centres need their actual status to have a complete picture of generation connected to the network.  This need will drive Distribution State Estimation which will may require synchronized measurements (synchrophasors) to implement due to the vastness and single-ended-ness of the distribution system.

 

FACTS, SVC and HVDC devices aim to provide solutions to some of the problems encountered in transmission networks such as mitigation of faults or enhanced transmission of power, but they also bring raise their own protection and automation issues and must be interfaced to the wider network.

Key to many of the solutions within the Network of the Future will be communications – the ability to gather information with standardized semantics from remote parts of the network and in order to make real time decisions based on an accurate knowledge of the state of the network. Many transmission substations already have access to adequate communication networks but many distribution substations have limited communication facilities and bring additional challenges to be able toif they were to be integrated them in the same way as transmission substations.

It is the WG’s belief that the standard IEC 61850 for “Communication networks and systems for Power Utility Automation” will feature prominently in the Protection and Control solutions employed for the Networks of the Future – especially its ability to provide self-description and generic configuration. Originally developed for use inside substations, it has been extended to cover substation-to-substation communication.  It also addresses sampled values of current and voltage in the process bus. This allows the realistic facilitates the use of new technologies such as non-conventional instrument transformers (NCIT).


 

  • Policy – Need author – environmental, incentives, usage policies, Grid connectivity Codes, ,communication requirements (e.g. – Italy requiring 61850 for all DER devices), Feed-in Tariffs , Time of use tariffs, virtual power plants /Demand Aggregation [MGA4](FIT), no more nuclear, etc.  Emilano – check with C6 WG

A feed-in-tariff (FIT) is a policy mechanism designed to accelerate investment in renewable energy technologies and facilitate preferential treatment by  ensuring long-term contracts to renewable energy producers, typically based on the cost of generation of the technologies. Under FIT, renewable electricity generators which broadly includes homeowners, business owners, farmers, as well as private investors are paid a cost-based price for the renewable electricity production [50]. Early reference of this mechanism appeared in ealry 2000 from Germany and several countries have followed suit since then. House-hold PV generation has received the maximum uptake following the preferential rates set using FIT mechanisms across several countries. With larger penetration of RES through such mechanisms has also resulted in focus around power quality and managing voltage rises and bi-directional flows in certain situations.  

23/10/2012[VL5]

A driver for change is the decrease of public´s acceptance on nuclear power plants. In Germany since the early 1970s electrical power supply from nuclear power plants have been a controversial issue, after the Chernobyl disaster even controversially. During the years the green movement established visions of a nuclear-free energy supply based on renewables however the German conservative-liberal government in that time adhere to nuclear power plants. In 1998 the social democratic green parties came to power pledging to phase-out the use of nuclear energy. In 2000 the social democratic green government reached an agreement with energy companies on the gradual shut down of Germany´s nineteen nuclear power plants by defining the remaining amount of energy a nuclear power plant is allowed to produce before being phased-out. It was expected to have a nuclear-free power supply after 2020. At the same time the government has started a development scheme to establish renewable energy sources. So far it is successful in that sense, that the portion of renewable energy sources on the final electrical energy consumption increased from 5% in 1998 to 20% in 2011.

The change of government in 2009 restore a conservative-liberal government to power, which passed one year later a law extending the operating lives of Germany´s remaining seventeen nuclear power plants by 8-14 years each. Despite of the extension of operation nuclear power plants have been considered as a “bridge-technology” in a socio-political consense and have to be replaced during Germany´s so called “energy transition”. Important aspects of Germany´s “energy transition” are:

  • Long term strategy till 2050 considering electrical power supply, heating and mobility
  • In 2050 80% of the final electrical energy consumption should be generated by renewable energy sources
  • Nuclear power plants are a “bridge-technology” to be completely replaced
  • Increased energy efficiency (the primary energy consumption in 2050 related to 2008 should be 50%)

Three months after the Fukushima nuclear disaster the conservative-liberal government revoke the extension of operation of nuclear power plants and phased-out 8 of 17 nuclear power plants in August 2011 enforced by public pressure. According to the actual circumstances the last nuclear power plant in Germany will be phased-out in 2022.

Considering Europe the tenor is more ambivalent. As Germany, Switzerland, Spain and Belgium decided the nuclear power phase-out, countries like Great Britain, France, Poland amongst others adhere to nuclear power plants and go for their expansion.

 

28/09/2012[VL6]

We can summarize a VPP as:

 

“Aggregation, control of a number of distributed generation units, grid connected and installed near loads. The aggregation control can be centralized or decentralized system supported by logic control algorithm and communication infrastructure, then treated as single large power plant”

 

The VPP is a system that controls the behaviour of a large amount of Local units (LUs). The LUs are both power-consuming and power-producing units (prosumers). The producing units could eg be small hydro plants, emergency gensets or wind turbines. The consuming units could  be cold storage facilities, greenhouses or drainage pumps. The important aspects of the units are that they have some flexibility in how they consume or produce energy. The VPP has to control the units in such a way that they support the energy system when power consumption and production are not in balance.

 

 

 

 

  • Market  - need author – future markets? By the hour? Interruptable? Micor-interruptable? Jorge, Nirmal

 

 

  • Social: Urban Grids, future social networking needs, use the grid as a back-up, desire for energy independence 

23/10/2012[VL7]

  • The future Energy Internet  ([05])will be concept is envisioned as a multi-layered large scale socio-technical system, in which the traditional distinction between producers, distributors and consumers of energy is replaced by the new role of prosumers, i.e., industries, cities, communities or individuals who can act both as producers and consumers of energy. Prosumers will become part of a global socio-ICT “ecology”, in which they can negotiate the energy they produce and consume. They will obtain direct financial benefits while promoting at the same time the growth of renewable energy sources. In alignment with this vision, the eNetworks concept tackles the Future Internet as a pervasive infrastructure ([06])  enabling the deployment of techno-social systems which have three dimensions physical (or ‘smart application’ dimension: smart power grid, transportation network, building infrastructures, computing facilities), cyber (the underlying large scale management ICT control infrastructure) and social (the users and their ability to form dynamic coalitions mediated via a communication network). Social networks are seen as an important mechanism for creation of such coalitions of prosumers.

 

 

5.     Vision of the Future Grid

Today, electric power is produced mainly by large geothermal, fossil, nuclear and hydro power plants at a limited number of centralized places. With the requirement for more environmentally friendly production (“green”, “sustainable”, “low carbon”, “carbon free”) and the related technical possibilities, the use of small, distributed plants (small hydro, gas, small CHP (local area), geothermal and photo-voltaic solar, wind, etc.) is increasing. In the next decade, more and more consumers will also be part-time real or virtual power producers by solar panels, micro-CHP (per home), electric cars, etc. and will join the trend towards decentralized power generation (see overview in Table 1b).

28/03/2012[VL9]

Table 1. Distribution generation capabilities and utility interface ([07])

 

Technologies

Typical Capability Ranges

Utility Interface

Solar/PV

A few to several hundred kW

dc to ac converter

Wind

A few hundred to a few MW

Asynchronous generator or inverter

Geothermal

A few hundred kW to few MW

Synchronous generator

Ocean

A few hundred kW to few MW

Four-quadr. Synchronous machine

ICE (*)

A few hundred to tens of MW

Synchronous generator or ac to ac converter

Combined cycle

A few tens of MW to several hundred MW

Synchronous generator

Combustion turbine

A few MW to fundred of MW

Synchronous generator

Micro turbine

A few tens of kW to a few MW

ac to ac converter

Fuel cells

A few tens of kW to a few tens of MW

dc to ac converter

(*) Internal combustion engine

 

 

 

 

 

 

 

 

Issues

Today

Future (next decade)

Power

generation

Centralized (fossil, nuclear and hydro)

Increasingly decentralized (high percentage of renewable sources)

flow direction

Power flow top-down

Power flow direction volatile and bi-directionnal

system 
structure

Large-scale overlapping voltage levels

Interconnected small-scale cells; large scale bulk power transfer

flow control

Passive by natural impedance

Active by FACTS and DGFACTS

Power system

information

Bulk data flow from the process level up to control centers

Information flow for the control centers according to their tasks

stability

Stability monitored e.g. by WAMS

Stability controlled by operators

Stability monitored and controlled
by automatics like WAMPAC

operation

Limited information and local actions

Comprehensive information and global but selective actions  → Smart Grid

Table 1b – Important issues in the evolution of the power system

 

 

Up to now, the power flow is directed from the generation to consumers through the transmission and distribution systems. With a lot of small, decentralized generation feeding in at distribution level, the power flow direction is periodically reversed. Therefore, not only the amount of power generation is volatile but also the power flow direction. In addition, the deregulation of the power market allows the consumers to decide individually about the acceptable price and, therefore, about the volatile amount of their consumption at any time.

 

The power system has a large scale structure with overlapping voltage levels. The backbone is the transmission grid which is reinforced today in some countries by UHV lines (≥ 1000 kV AC) and/or long HVDC lines. Transmission limits and distributed energy production have resulted in the concept of local power system cells, trying to balance out production and consumption locally, but with the capability to import or export electric power on demand. 

 

The power flow in lines and cables is limited by their natural impedances. Active elements named FACTS (flexible AC transmission systems) and DGFACTS (low power FACTS) allow the control of the impedance and, therefore, the power flow over a reasonable large range.    

 

The above mentioned change in the use of the power system is a big challenge for the operation of the power system itself. The most important issue is to keep a good stability margin for the power system and to avoid black-outs. For this purpose, not only much more information from everywhere in the power system is needed, but also coordinated actions in much more control points in the power system. All these requirements request fast, adaptive and selective response to cope with any situation by appropriate automatic actions. Thus, although the term “smart grid” has been introduced with main focus in the distribution level, such smartness is needed in all levels, and may be seen as a collective property of the power system. As important side effect, this will blur the differences between transmission and distribution.

Add picture of Grid Hierarchy with a description of the various levels[MGA10]

 

5.1     Bulk Transfer and Super Grids

5.1.1     Definition

Japan Reneawable Energy Foundation proposes to build Asia Super Grid, which overlays HVDC interconnectors on the existing bulk power transmission systems. One of the purposes is to incorporate more reneawable energy into the grid. If the super grid is built as proposed, the total of 35,000 km HVDC interconnectors will surely pose challenges to the protection of the interconnected systems.

 

Bulk transfer of power designates a situation where the load centers are geographically distant from the points of power injection. This situation may be permanent or show seasonal or daily variations. A typical example is the production of wind energy in the northern parts of Europe, which has to be transferred to the load centers situated more in the south.

Bulk transfer can take place within one grid or between to non-synchronous grids. In this latter case, HVDC links are normally used. In this context, the notion of a supergrid has emerged. It designates a superposed bulk transfer grid connecting one ore several grids in multiple points, close to the load centers. This type of grid is associated to huge bulk power transfer between high-capacity injection of power produced by renewable energy sources (wind, solar, wave, geothermal) and load centers.

[VL12]

  • Connects multiple grids or coordination councils
  • Trend to DC (underground pressed by policy) / Dispatchable
  • Power transfer to balance the local diversity and time variability of renewable energy resources (Offshore-Windfarm, hydroelectric power plants, pump storage power station, solar energy electric power station, tidal energy, ocean currents, ocean thermal generation) each other and in relation to load centers[MGA13]
  • Will drive the need for new protection strategies and control technologies and to maintain stability – will be predicated on the generation mix (high vs. low inertia systems)

 

  • Super-conducting out of our 20 year range

 

5.1.2     Example of South African supergrid– Adam

The Super Grid may have a different purpose on different power systems depending on location of bulk energy sources, distances between load and energy centres, location of storage facilities, amount of power to be transferred, etc. therefore few examples from various utilities and shown below.

 

The Super Grid in South Africa is not completed yet and is currently under construction at the 765kV AC voltage level with possible injections of HVDC from remote, mainly large hydro power stations in neighboring countries up north.  One such HVDC injection from 1500km distant hydro power station in Mozambique at 533kV voltage level is in operation for many years and more are envisaged in future.

 

The purpose of Supergrid in South Africa is to improve power transfer capabilities of corridors between generation and load centres, improve overall reliability of the network in cases where local generation may not be available as a result of major incident, improve voltage profiles as well as to facilitate new developing load centres and prospective locations for new power plants including large number of geographically spread renewable energy projects.

 

 

 

 

 

 

 

 

 

 

Figure 5.1.1.1.  765kV AC / HVDC Supergrid example from South Africa.

 

On the above figure the 765kV Supergrid (purple) overlays the 400kV backbone transmission network (green).  The existing 533kV HVDC is indicated in red.  Possible future HVDC injections are not shown on the above figure due to uncertainty of exact injection points.

 

The interconnections of South African network with 9 neighboring countries to form Southern African Power Pool (SAPP) can also be considered a Supergrid in a context of relatively small amount of power transfer over large distances at EHV level.  The SAPP interconnections operate at 400, 330 and 220kV AC as well as 350 and 500kV DC.

 

5.1.3     Links and DC Networks - Volker

Although Current Source Converters have been the foundation of existing DC links, the availability of Voltage Source Converter technology based on IGBT/GTO devices for DC links with a huge power transit (up to 1000 MW per link is reported) enables the DC connection of offshore wind farms and islanded networks with small short circuit power to transmission networks. The number of commissioned and planned point-to-point links of this type is increasing, partly as result of the increase of offshore wind production which has to be connected to the transmission network. A more recent trend is the insertion of HVDC links in existing AC networks, mainly due to constraints of construction of new overhead lines in some regions. New conventional, thyristor-based Line Commutated Converter (LCC) DC links are also being commissioned. They have now been used for decades to interconnect large, non-synchronous AC transmission systems.

It is anticipated that Future networks will need an increased part of HVDC DC lines and other VHV bulk transmission lines in order to enable the transport of renewable energies (offshore wind farms, PV generation) to the load centres.

 

5.1.4     Multi-terminal DC lines and DC busses – Volker

The next logical step will after the implementation of isolated HVDC links will be the creation of multi-terminal DC lines and / or HVDC networks with HVDC busbars. Faults appearing in this network will have to be eliminated in a selective way, as it is done today in AC networks, or the complete DC network will shut down.

In conventional HVDC links, faults on the DC part are eliminated by the HVDC pole controllers by stopping the commutation process. This leads to a disconnection of the HVDC sources in a very short time (some ms) and to a de-energisation of the DC line in some seconds, depending on its time constant. The short reaction time of the DC poles, which is related to the thermal capability of the valves, imposes a very short time (some ms) for a selective elimination of an DC fault (some ms), compared to the time in AC networks (some 10ms).

Selective elimination of faults in DC networks will require :

  • Fast protections capable of identifying the faulted section in some ms.
  • Fast DC circuit breakers capable of separating the faulted section from the rest of the network with an operating time in the range of some ms.

Although under investigation, none of these devices are available today.

 

AC protection algorithms based on steady-state fault impedance estimation cannot be used because here because of the time constant of DC networks. It seems clear that DC protection will have to be based on the transient phenomena observed after DC fault inception.

 

Conventional AC circuit breakers take advantage of the fact that the current becomes zero twice every cycle. In DC, fault current can be expected to increase and then to decrease, following the different time constants of the network and depending on the fault location. A fast HV DC circuit breaker must thus be capable of interrupting a direct, inductive current in some ms.

 

It is thus clear that the protection- and fault elimination systems in DC network will be very different from those in operation in today's AC networks. It can also be said that they have to meet requirements which seems to be much more constraint than those for their AC counterparts. When these equipments are will be developed, it can be anticipated that the algorithms, approaches and technologies employed may influence the design of their AC counterparts leading to a shorter fault elimination time also in AC networks.

5.2     Regional sub-transmission grid  - Adam

5.2.1     Definition

A regional sub-transmission grid has the following characteristics:

The subtransmission network for the purpose of this report can be illustrated on the following diagram.  [VL14]

 

 

 

 

 

Figure 5.2.1.1.  Illustration of typical subtransmission topology.

The subtransmission network operates in ring or meshed topology and connects high voltage distribution networks to transmission or super grid systems and as such provides limited transmission capabilities.  Subtransmission systems usually operate at voltage levels in the range from 66kV 50 kV to 132kV150 kV.

 Depending on the utility, the protection systems may not be fully redundant and only incorporate one main protection relay required to provide instantaneous fault clearance and back-up protection, overcurrent and earthfault typically.[MGA15][MGA16]

 

5.2.2     Subtransmission aspects in future networks - Adam

Distributed energy sources are likely to be connected to lower voltage networks due to their size and location more specific to energy sources than availability of electrical networks.

Currently the DER connections are driven by Grid Codes which stipulate specific requirements for connected energy sources and protection of the grid.  To meet these requirements, however, protection systems on many distribution networks will have to be upgraded.  On subtransmission interconnections possibility of bi-directional flow of energy, synchronisation and overvoltage is normally catered for and less additional equipment may be required.  The possibility of sustained islanded operation as a result of connected DERs and reliable detection of such operation will have to be provided to ensure safe operation of the network. 

Concepts of safe and stable micro grid operation also may become applicable to subtransmission level connected DERs.  In addition, to ensure adequate protection performance, the specific contributions to fault current of connected DER must be known, modelled and analysed.  As indicated in chapter x.x.x. behaviour of each technology (synchronous machine, induction generator, DFIG or full converter) is different during short circuits and in some cases symmetrical components models are not available.  To analyse such systems detailed time domain models (both RMS and EMT) must be available, which used to be a problem to date.  Similarly to evaluate protection performance and to calculate its settings time domain simulations may become necessary.  Methods based on time domain simulations have been already implemented in power utilities where dynamic performance of the network has significant impact on relay settings, mainly on transmission systems.

Many distribution networks operating as radial supplies or with predetermined open points may become ring connected to support transmission systems in situations of transmission capacity shortages or due to outages.  Future smart grids may require automation of this process by closing of open points with automatic and intelligent (based on current needs and measurements) load shedding to sustain itself and minimise shed load if necessary (self-healing concept) as described in chapter x.x.x.[MGA17]

Automatic separation of underlying subtransmission connections due to overload as a result of unavailability or tripping of overlaying transmission connections may also be mitigated by automatic load shedding and restoration systems.

Protection systems on such subtransmission networks will have to become adaptive or immune to different operating regimes due to possible wide range of fault current contribution conditions.  Usually the subtransmission networks are equipped with impedance or differential protection relays but co-ordination of back-up functions can become compromised.  The impact of dynamically changing voltages and currents on directionality determination also needs to be evaluated.  The adaptive criteria could be based on local or remote measurements or entirely controlled by centralised, for a particular network, “master controller” using reliable SCADA or state estimation information about status and loading of the network.  In addition effective back-up mechanisms would have to be developed to minimise network disruptions in case of controller or communication failures.

With the advent of multipoint HVDC networks their applications to subtransmission networks may become an attractive alternative to increase subtransmission capacity using existing infrastructure (e.g. conversion of AC to DC) and manage power flow.  Parallel operation of AC and DC at subtransmission voltage levels should be well analysed from voltage and transient stability perspectives depending on specific system configuration, location of generation and its characteristics.  Possibility of improved voltage control using VSC technology can also be explored for additional benefits.

Methods to analyse possible AC to DC faults with associated transients will have to be developed and become part of electrical engineers’ curriculum.

Switching on “future networks” may become much more frequent while self-healing and adaptive protection concepts are fully utilised.  More attention may be required to techniques aimed at minimising switching transients. 

 

5.3     Distribution grid Mika

Definition:

 

Distribution grid connects the transmission and sub-transmission grid to the customer level. It enables the connection of the huge amount of low and medium sized loads. Now and in the future the distribution grid also receives the increasing amount of the distributed medium and low level generation infeed. Distribution grids are usually owned by Distribution System Operators (DSO’s). The operating area of the DSO’s can cover in some case as much as the whole country, some munipalicities or only a one municipality depending on the country and continent. Distribution grids have usually at least two voltage levels, medium voltage level and low voltage level. In Europe the medium voltage level is typically between 1-70 kV and low voltage level is worldwide under 1 kV.

 

Protection in the distribution grid level is not characterized by the network stability needs but the selectivity of the protection, thermal endurance of devices and safety issues. Also the increasing amount of distributed generation has brought demands to the DSO’ protection to take into account fault currents and back-voltages produced by the DER-units

 

By this definition the distribution grid is subjected to structural changes due to an increasing penetration with DER-units. The actual passive top-down network structure changes into an active network structure with DER-infeeds. These changes have a physical as well as technical character and address network topology, DER connection, network dynamic characteristics and contribution to ancillary services.

 

The increasing need to reduce the outage quantity and duration of electricity customers have brought pressure to the DSO’s to make investments to their distribution grids. Most of the customer outage time is caused by the faults in the medium voltage network. DSO’s will use different kind of methods to improve their reliability of electricity supply, mainly in the MV level. For example DSO’s can use methods listed below:

  • cabling of overhead lines to avoid faults caused by climate conditions
  • improving network strcture from radial to loop and even to closed loop solution reducing the outage times and amount of customer interruptions
  • adding network automation to reduce outage times and even to implement automatic voltage restoration

 

5.3.1     Network Structure  – Volker

There are two main topologies in electrical networks :

  • Meshed networks
  • Radial networks

 

Radial networks are mainly used in for LV and MV distribution networks. They are characterized by a topology providing one single point of connection to the bulk network. Usually, a busbar is connected by via a transformer to the upper voltage level. Several outgoing feeders are connected to this busbar, usually by a circuit breaker per feeder. In particular in rural areas, these feeders can supply several second- or third level branches.

This structure is simple to design and cheaper to build and also greatly simplifies feeder protection, since faults appearing in an outgoing feeder can be selectively cleared by protection simple schemes, mostly based on time-delayed overcurrent relays or fuses.

In many cases, open-loop topologies are used in order to provide alternate supply paths in case of outage of a feeder section. In any case, a fault on a radial system leads to an outage for the customers fed by the faulted feeder and to a voltage sag for all customers supplied by the same busbar. Autoreclosing schemes and, in case of permenant faults, modification of feeder topology allows to restore supply. The time required to restore the supply for a given customers depends on many parameters, including his position on the feeder with respect to the fault, availability of alternate supply paths, nature of the fault and availability of remote operated switches.

The radial structure is designed to supply passive loads and is less adapted for the connection of power producers. If generation is connected to radial feeders, issues like protection selectivity, coordination of recloser actions, islanding detection and – operation arise. These issues become more difficult to address with a growing amount of power generation in the radial network.

 

The topology of meshed networks is characterized by loops which may provide several links, including transformers to the upper voltage levels, between two given busbars. These networks are usually designed and operated according to the N-1 principle, where the outage of one single element of the networks does not disrupt the connection of any given busbar.

Meshed networks are thus more robust, but also more expensive to design and to built. They require more evolved protection schemes like distance or differential protections in order to guarantee a selective elimination of faults. Depending on the voltage level, protections are doubled in order to provide redundancy in addition to backup protection schemes.

Meshed networks are designed for power injection at various point of the network and the load flow in a given line may change frequently without affecting the performance of the protection systems. Meshed networks can also contain single ended tie lines to connect load or generation and they may be operated in open loop topology depending on the network conditions.

 

In addition to meshed and radial topologies the type of the distribution lines of the network is important. There are cabled networks, mixed networks (with cables and overhead lines) and pure overhead line networks. Cabled networks are used normally in the city areas and population centers. Overhead lines are used in the rural areas but also in many countries in the populated areas. In the Northern Europe some utilities have started programs for cabling also some rural medium voltage distribution networks because of long and large outages caused by masses of falling trees during storms.

 

5.4     MicroGrids – Mark, Tianshu

  • Definition
  • Mixture of machine and inverter based generation
  • Capable of stand-alone operation
  • Summary[MGA19]
  •  

5.5     Local Energy Networks  [n20]- Paul Mydra

The Local Energy Network facilitates the functionality of the ElectriNetSM.  Overall, the combination allows for the operation of a power system that is:

  • Smart, self-sensing, secure, self-correcting and self-healing
  • Able to sustain failure of individual components without interrupting service
  • Able to focus on regional, area-specific needs
  • Able to meet consumer needs at a reasonable cost with minimal resource utilization and minimal environmental impact
  • Able to enhance the quality of life and improve economic productivity

5.5.1     Design Criteria

The design criteria that would be employed to meet these conditions must address the following key power system components or parameters:

  • End-use energy service devices:  The end-use devices are the starting point in the design of the Local Energy Network.  They are the point of interface with the energy user and the mechanism by which the energy user receives the desired service, such as illumination, hot water, comfortable space or personal temperature (heating or cooling), and entertainment.
  • System configuration and asset management:  Advanced technologies, such as high-power electronic-based controllers can provide unprecedented flexibility and speed in controlling the flow of power over transmission lines, while new energy storage units can help level loads and improve system stability.  Innovative technologies can increase the amount of power that can be carried along existing rights of way.
  • System monitoring and control:  Very rapid declines in the cost of sensors and communications and continuous advancements in those areas are opening new opportunities to increase the use of existing transmission and distribution facilities by monitoring and controlling the operation of numerous devices simultaneously.
  • Resource adequacy:  Integrated asset management would be a requirement to realize the ElectriNetSM.  Advanced Central Generation combined with small power generation and storage devices and direct control of end-use devices would result in a robust, flexible system.
  • Operations:  A variety of issues need to be addressed to develop a new operational paradigm that guarantees optimum operation of the electric service system at all times.
  • Connectivity:  One of the highest priorities in developing a perfect power system is to develop system architecture for the communications, data networking, and robust control infrastructure required to support a smart, self-sensing, secure, self-correcting and self-healing system and to provide energy users with choice in energy services.  In work already underway at EPRI, in an effort called the IntelliGrid, an integrated energy and communications systems architecture is envisioned that will build on secure open systems specifications to define the capabilities and the functional requirements for the system.

Local Energy Networks increase the independence, flexibility and intelligence for optimization of energy use and energy management at the local level; and then integrate Local Energy Networks to the Smart Grid.  Local Energy Networks, energy sources and a power distribution infrastructure are integrated at the local level.  This could be an industrial facility, a commercial building, a campus of buildings, or a residential neighborhood (refer to Figure ___).  Local area networks are interconnected with different localized systems to take advantage of power generation and storage through the Smart Grid enabling complete integration of the power system across wide areas.  Localized energy networks can accommodate increasing consumer demands for independence, convenience, appearance, environmentally friendly service and cost control.

While some of these Local Energy Networks can operate in a stand-along mode, integration into the distribution system allows interconnection and integration with technologies that ultimately enable a national Smart Grid.  Inherently, Local Energy Networks can operate somewhat independently, but their value is maximized when they are nested within the Smart Grid.  This nesting concept also allows for increased overall stability within the power system.  In various configurations, storage (large and small) and power electronics (large and small) at all levels of the power system can be utilized to reduce interdependencies between system components and make the system immune to temporary disturbances.

Local Energy Networks will require increased investment in local intelligence and infrastructure.  In turn, integration of Local Energy Networks will require higher levels of integration involving more significant infrastructure transformation in communications and control, as well as in the overall power delivery infrastructure.

The availability of comparatively inexpensive and clean Central Generation and storage (e.g., advanced coal, advanced nuclear, advanced hydro and advanced large wind systems) will occur in parallel with the development of more localized or distributed infrastructures.

5.5.2     Components of Local Energy Networks

Local Energy Networks integrate energy sources and a power distribution infrastructure at the local level.  This could be an industrial facility, a commercial building, a campus of buildings, or a residential neighborhood.  Local Energy Networks allow for:

  • The optimization of energy availability across a larger variety of energy sources, resulting in improved economics.
  • The creation of an infrastructure for more optimum management of overall energy requirements (heating, cooling, and power).
  • The control and management of reliability at the local level.
  • Technologies needed to facilitate Local Energy Networks include:
  • Energy-Efficiency Appliances and Devices – End-use equipment will have to become very efficient in order to optimize Local Energy Networks.  Losses have dramatic impacts on the thermal design requirements of equipment and result in the need for larger electrical storage systems, as well as energy sources.  Many energy-efficiency gains will be achieved through miniaturization of technologies (and associated reduced energy requirements for operation), as well as advancements in power electronics (see below).
  • Energy Storage – As with portable power systems, energy storage (including thermal storage) is a key to the success and reliability of localized power systems.  In this case, the focus of energy storage can be larger systems that are part of the overall local energy system as opposed to the emphasis on energy storage in each device.  Energy storage systems, sized for 1 to 15 kW and upwards to larger systems of MW capacity for industrial and large campus applications, are needed to enable the localized power systems.  The single biggest obstacle for this application is the high cost of energy storage.
  • Distributed Generation – Photovoltaics, microturbines and fuel cells can become essential components of Local Energy Networks.  Storage systems will provide the flexibility to utilize renewable energy sources as appropriate.  Development of these technologies has been ongoing for many years, but there are still tremendous opportunities for advancement of the technologies.
  • Power Electronics – Local Energy Networks will take advantage of local direct current (DC) distribution with inverters to provide alternating current (AC) as required by motors, etc.  Power conversion technologies to achieve this will dramatically improve the energy efficiency and reliability of the local systems.
  • Sensors – The integration of different energy systems and end-use devices will require development of low-cost, highly functional sensors to track performance and status of the different parameters and components.  These will be embedded sensors in devices that provide information about status, energy requirements (present and future), and problems.  The sensors must have integrated communications capability for interface with the overall facility energy management.
  • Building Systems – Energy management moves from the requirements of individual devices to optimization at the local facility level.  Energy management functions must optimize heating, cooling and power requirements with available energy sources and storage systems.  This optimization function (especially including the storage capability of individual loads and the overall system) can provide tremendous benefits in terms of using renewable energy sources in an optimum manner and improving the reliability of the overall system.  Additionally, combined heat and power applications are important to the viability of localized power systems.  Combined heat and power systems have significant benefits in that the fuel is used more effectively.  For instance, the car could be used to provide both, heat and power to a home, while parked in the garage.  The heat could also be used to provide air-conditioning in buildings using chillers that convert the waste heat to cooling.
  •  

1)     Electrochemical Batteries

Electrochemical batteries store the energy in the form of charged ions. It consists of two or more electrochemical cells. The more common ones include lead-acid, nickel-cadmium (NiCad), lithium-ion (Li-ion), sodium/sulfur (Na/S), zinc/bromine (Zn/Br), vanadium-redox, nickel-metal hydride (Ni-MH) etc.

a)      Flow Batteries

Some electrochemical batteries store the electrolyte in a separate container (e.g. a tank) outside of the battery cell container; unlike automobile batteries are called flow batteries.  Vanadium redox and Zn/Br are two of the more familiar types of flow batteries. Expandability is the key advantage to flow batteries is that the storage system’s discharge duration can be increased by adding more electrolytes. The maintenance and replacement of electrolyte is also easier.

b)     Capacitors

An array of capacitors can be used to store electrical energy in form of electrostatic charge. A large number of capacitors in the array could be used for energy storage applications. The only limitation to these storage is the short discharge time, thus they are suitable for short or frequent discharge cycles to store significant amount of energy.

c)      Compressed Air Energy Storage

If electric energy could be stored in form of pressure that could later be used to generate electrical energy. Such storage is called Compressed Air Energy Storage (CAES) involves compressing air using inexpensive energy to be used later when the energy is worth more. The compressed air is heated and released into turbine generator system to generate electricity. Larger CAES plants have large underground storages located on geological formations, such as salt formations, aquifers, and depleted natural gas fields. Tanks or on-site pipes are used for smaller CAES plants mostly above ground.

d)     Flywheel Energy Storage

Another form of storing the electrical energy is to convert it into kinetic energy. A possible option is a Flywheel electric energy storage system. In this system, electrical energy is converted into kinetic energy in the form of rotating flywheel, which could again be converted back into electrical energy by using electrical generators through slowing down the flywheel.

e)       Pumped Hydroelectric

Electricity could also be stored in the form of potential energy. When required, it could be converted back to electrical energy again. Such a system is possible through pumping water to a higher reservoir through electric motors and pump set. Using turbine and generator set the potential energy stored in higher reservoir in form of water level could again be converted back to electricity. The water is later released when energy is more valuable.

f)       Superconducting Magnetic Energy Storage

Electrical energy could also be stored in form of magnetic energy. The energy stored could die out due to losses. However, superconductivity could be achieved through cooling down the coil to a temperature lower than required for superconductivity. Such systems are called Superconducting Magnetic Energy Storage (SMES) This system consists of a coil made of superconducting material. Additional SMES system components include power conditioning equipment and a cryogenically cooled refrigeration system.

g)      Thermal Energy Storage

  • There are various ways to store thermal energy. One way is to make ice when electricity is available and then ice could be used to avoid electricity usage for cooling purposes. New modern ways are also storing energy in phase change materials to avoid heating and cooling needs in building [12]. There are about 17 energy storage applications which could be categorized as shown in Table YYZ.

 

TABLE YYZ

Energy Storage Applications

Electric Supply

Ancillary Services

Grid Systems

End User

Renewable Integrations

Electric Energy Time-Shift

Load Following

Transmission Support

Time-of-use (TOU)

Renewables Energy Time-shift

Electric Supply Capacity

Area Regulation

Transmission Congestion Relief

Demand Charge Management

Renewables Capacity Firming

 

Electric Supply Reserve Capacity

Upgrade Deferral

Service Reliability

Wind Generation Grid Integration

 

Voltage Support

Substation On-site Power

Service Power Quality

 

 

6.     Connection of Various Energy Sources[VL23]

 

Because of the wide diversity of regenerative energy resources a wide diversification and variety of power generation technologies are available in future distribution grids. The spectrum spans from directly coupled synchronous or asynchronous machines to machines coupled to the distribution grid with converters (refer to

Table 2Table 2). Thereby different infeed characteristics regarding loading and short-circuit behavior occur.

 

The point of common coupling (PCC) of DER-infeeds (typically up to 10 MW) to the distribution grid is the result of a detailed economical and technical investigation of each single connection demand under consideration of valid guidelinies, rules, etc. In abstract terms the PCC is undefined, that is DER-infeeds [MGA24]can be located everywhere in the distribution network and their dimension is limited by the valid guidelines, rules, etc.

 

 

Table 2. Power generation and coupling technologies of different types of power plants (PV: photovoltaic, FC: fuel cell, WP: windpower, HP: hydroelectric power, CHP: combined heat and power, MT: microturbines)

 

Table xy

Generation

Coupling to Network

Type of Energy Source

PV

FC

WP

HP

CHP

MT

BAT

DC-Source

DC-AC Inverter

X

X

 

 

 

 

X

AC-Sourced

Converter (back-to-back DC)

 

 

X

 

X

X

 

Synchronous machine

Transformer/direct

 

 

 

X

X

 

 

Asynchronous machine

Transformer/direct

 

 

X

X

X

 

 

Doubly Fed Induction Generation (DFIG)

Converter/Transformer

 

 

X

 

 

 

 

 

6.1     Connection of Distributed Energy Resources to Distribution

 

 

  •  

DER Connection / Dis-connection  Tianshu – add as part of 6.5.1

 

28/09/2012[VL26]

The use of distributed energy resources has increased during last few decades. Due to increasing demand to power supply reliability, some environmental and economical concerns further decentralization of power systems should be expected. The connection of DG to the distribution network changes power system condition and can have a serious impact on relay protection operation.

 

Fig. 1 shows simple radial distribution network with DG connected at the substation 2 ([08]).. Dependence of short circuit current on fault location is presented in fig. 2. When the fault occurs at the line 2 or line 3 total short circuit current is increased due to the connection of DG, however contribution from bulk grid resources is reduced. The same effect appears itself if there is fault at the line 1 over transient resistance. When fault occurs at the adjacent feeder (Line 4) reversed short circuit current is provided by DG.

Impact of DG can lead to:

loss of protection sensitivity;

increase of fault clearing time (particularly due to cascading fault trips);

selectively problems;

excessive trips.

 

Nav_01

 

Figure 1 Radial network with distributed generation

 

 

As it was shown above integration of DG in power system impacts on fault currents directions and levels and can cause protection maloperations. Moreover operating conditions of DG and its fault current contribution continuously changes that make protection development more complicated.

 

Nav_02

 

Figure 2 Dependence of short circuit current on fault location

 

Protection of medium voltage distribution network is often based on non-directional overcurrent relays. In case of using definite time characteristic fault clearing time can be not acceptable and leads to stability problems. Inverse time-current characteristics are more preferable from this point of view. However their setting requires careful consideration of all possible DG conditions.

Fig. 3 shows dependence of fault clearing time on fault location for distribution network described above without and with DG (very inverse time-current characteristic). The connection of DG leads to increase of fault clearing time of line 1 protective relay in the case of fault locations at the lines 2 or 3, reduce difference in clearing time of line 2 and line 3 protective relays. To provide proper functioning of relay protection for this case relay setting should be adjusted:

normal inverse time-current characteristics should be used that allow to reduce operation time for line 1 protection (in backup mode);

– required clearing time difference of line 1 and line 2 relay should be provided for islanded operation of DG;

required clearing time difference of line 2 and line 3 relay should be provided for interconnected operation of DG.

In some cases using inverse time-current characteristics allows to provide protection selectively when the fault occurs at the adjacent feeder.

 

 

Nav_03 

Figure 3 Dependence of fault clearing time on fault location (very inverse characteristic):

a) without DG, b) in presence of DG

 

 

 Nav_04

Figure 4 Dependence of fault clearing time on fault location (normal inverse characteristic):

a) without DG, b) in presence of DG

 

In general the impact of DG depends on DG capacity and location, configuration of distribution network. In case of relatively small DG impact common protection solution for medium voltage distribution network is still acceptable but requires relay settings adjustment; if DG impact leads to serious sensitivity and selectively problems changes of protection principles is required. For instance good results can be achieved by applying protection principles inherent to high voltage networks (distance protection, protection with communications links) with some adaptation of protection characteristics.

 

6.2     Implementation of Tap Changers in Distribution – Mika
 

6.3     Connection of various Energy Resources to the transmission network[VL27]

 

 

RES in electrical power system is likely to increase significantly to accommodate future load growth, maintenance of the system security and addressing environmental considerations. The increasing penetration of renewables has technical and economic impacts to be considered by system planner and operators. 

 

This is likely to be witnessed globally but here we present as an exemplar the case study of New Zealand which electrically has two 220 kV power system areas interconnected through a 1200 MW HVDC link. Additionally it is to be also noted that New Zealand Power systems operates under an wholesale electricity market operation since 1996.

 

The New Zealand power system has a peak demand of 3500 MW in the south island and 4500 MW in the north island; thus a total of 8000MW.  These two islands are connected through a 1200 MW HVDC link.  The energy produced to cater the energy need by different kinds of generations from 1990 to 2010 is shown in Table 6.3A and Fig 6.3A

TABLE 6.3A

  ANNUAL ELECTRICITY GENERATION, GWh

Year

Hydro

Thermal

Geothermal

Wind

Total

1990

22953

5956

2049

0

31467

1995

27259

5426

2049

1

35244

2000

24387

10474

2756

119

38285

2005

23094

14286

2981

608

41514

2010

24470

11140

5551

1618

43401

 

 

 

Fig. 6.3A . Typical New Zealand Generation Mix

 

 

 

This mix shows that about 78% are of renewable type generators currently. The NZ government has the renewable target of 90% that needs to be achieved by 2025. To meet this target there is significant additional wind plant and geo-thermal plants under construction or planned for the near future.

 

The transmission system operator, Transpower, maintains the grid that consists of 220kV, 110Kv, 66 kV HVAC networks and the inter-island 1200 MW HVDC links. Presently, the entire grid consists of 11,800 km of lines and 178 substations and switchyards, which are owned and operated by Transpower New Zealand Limited. However, majority of its assets are older than 40 years and several are more than 70 years of service.

 

 

In the New Zealand Electricity Market (NZEM), operated by Transpower as an Independent System Operator (ISO), sufficient contingency reserve generation in the system is maintained so that these reserves can be called on and used to cover the outage of the “risk generators”.  The risk contingencies are well defined, and the amount of reserve needed to cover such contingencies is determined dynamically using energy and reserve co-optimisation. The generators can offer same generation capacity from a unit into both energy as well as reserve markets at the same time interval.  In addition frequency regulating reserves and black start capabilities are procured separately through tendering process, outside optimization.

The generators are self committed through their offers and must submit their generation-price schedules before the gate closure of 2 hours before real time operation.  The forecasted demand (and demand bids used only in pre-dispatch schedule) is used in the market optimizations before real time operation.  This is a ½ hourly nodal spot price market, but it also has 5 minutes dispatch and pricing schedules.  Final prices are settled after collection of meter data, next day.

 

    At the centre of the dispatch procedure is the “Scheduling, Pricing and dispatch tool (SPD)”. Market is cleared using the SPD.The characteristics of the New Zealand Electricity Market reveal that it is “Energy Only” market, which means generators do not get any incentive to maintain system capacity besides energy charges. They receive payments based on amount of energy delivered, so they try to maximize the energy charges. Participating in Fast Instantaneous Reserve (FIR) and Sustained Instantaneous Reserve (SIR) are other significant incentives which a supplier could receive in form of payment to maintain the reserves.

New Zealand Electricity Market has a key challenge of capacity. New Zealand electricity generation is mostly hydro dominated, with high initial cost and low operating cost throughout the year. In times of dry season or peak demands, generation from other sources are required such as thermal and especially from fuel plants in order to meet the demand. As the electricity cost from fuel plants is high so it increases the spot prices in market. Due to high operating costs of fuel plants and non-utilization of these plants under presence of cheap electricity from hydro plants, investors get discouraged. In order to cater the capacity issues, there are two options; either additional generation sources, demand side option or energy storage options. Due to such situation in market energy storage may become very attractive for future needs.. 

 

Reliability of electric supply is also a growing challenge an it is equally important for citizens and policy makers especially in this digital world where, power outages are almost unacceptable. When Power generation is dominated by hydro-electricity and with a fact of having limited capacity to store water, puts electricity industry at risk of supply shortages in dry years. Other renewable energy options such as wind, geothermal, solar and energy storage plants could possibly contribute towards reliability but how much; this still needs to be investigated.

 

Modern power system operation is a high-order multivariable process whose dynamic response is influenced by a wide array of devices with different characteristics and response rates. Depending on the network topology, system operating condition and the nature of disturbance, different sets of opposing forces may experience sustained imbalance leading to different forms of instability. Power system stability can be classified under voltage, rotor angle and frequency stability.  Security is determined by system operating state. 

 

Power system's operation is governed by three set of equations. A differential set of equations representing the dynamic behaviour of the system.  An algebraic set comprising equality constraints, which refer to the generation-load system balance and a second set of algebraic equations comprising inequality constraints which states that all network elements operate within their bus voltage and power limits. From a reliability analysis framework the system can be classified under the following operational states Normal, Alert, Emergency, In extremis and Restorative.  In ‘Normal state the balance between generation and load along with losses is met with no equipment being overloaded. In ‘Alert State all the constraints are satisfied but the reserve margins are not enough to guarantee that the constraints will remain satisfied when a severe contingency occurs. In this state preventive action can be taken to restore the system to the normal state.Emergency state where some of the inequality constraints are violated, which means that some network equipment is overloaded.In extremis state both equality and inequality constraints are violated. During ‘Restorative state the operator performs control actions in order to reconnect all the facilities and to restore all system loads. The system can reach either the normal or the alert state, depending on the conditions.

There are three levels of security assessment:  Security monitoringif the operating conditions are satisfied. Security analysis whether system is secure to withstand a disturbance without getting into an emergency state else it is insecure or in an alert state. In ‘Security margin determination the operator tries to find how much load or transfer increase can be accepted without leading to insecure state. Security assessment can be classified into Static Security Assessment (SSA) and Dynamic Security Assessment (DSA) according to the differences in the required analysis methods.

 

Static security Assessment (SSA): are methodologies that verify bus voltage and line power flow limits for the post contingency operating state.  Dynamic Security Assessment (DSA): evaluating the stability and quality of the transient processes. In this case, DSA aims at ensuring that the system will be stable after the contingency occurrence and that the transients caused by such a contingency will be well damped, of small amplitude and with little impact on the quality of service.

 

The state of art on-line DSA installations around the world is listed in the Table XXX.

The Transmission System Operator (TSO) uses industry standard EMS Contingency Analysis programs to analyze static security. In this analysis every credible contingency is analyzed for the components within SO’s oversight. Overloads on lines and transformers, over-voltage and under-voltage conditions, as well as maximum angle difference are all monitored during the contingency analysis. Some critical n-2 contingencies are also examined where these are considered credible contingencies.

 

 

 

TABLE  XXX

SOME ON-LINE DSA INSTALLATIONS AROUND THE WORLD

Country/Region

Location/Company/Projection

SCOPE

Status

 

 

TSA

VSA

SSSA

FSA

Australia

NEMMCO

 

I/S

Brazil

NOS

 

 

I/S

Canada

Hydro-Quebec

 

I/S

China

Beijing Electric Power Corp

 

 

 

I/S

China

CEPRI

 

 

 

I/S

Europe

Omases Project

 

O/S

Greece

Hellenic Power System

 

 

I/S

Ireland

ESB

 

 

I/S

Japan

TEPCO

 

 

I/S

Malaysia

Teneaga National Berchad

 

 

I/S

NewZealand

Transpower

 

I/S

Russia

Unified Electric Power System

 

 

I/S

South Africa

ESKOM

 

 

I/S

USA

Southern Company

 

 

 

I/S

USA

Northern Company

 

 

 

I/S

 

Legend:

TSA Transient Security Assessment

VSA Voltage Security Assessment

SSSA Small Signal Security Assessment

FSA Frequency Security Assessment

I/S In Service

            O/S Out of Service

 

With larger penetration of renewable resources into the transmission network DSA will have to be reassessed within the context of the changing generation mix.  

6.4     Network Dynamic Characteristics

  •  Network Dynamic Characteristicsas created by the addition of Non-conventional Energy sources – add as 5.5.3 – need an author (include variability of wind on power oscillations include in 6.3.7)
  • Short circuit[MGA28]- Nirmal
  • Frequency tracking

 

Several types of Wind Turbine Generators have been defined (cf. Appendix 2). Figure xy below shohs as example WTG type IV.

Figure xy : Type IV Wind Turbine Generator

Fig. 4. Type-4 Wind Turbine Generator Nirmal NZ

[n29]

 

6.5     Effect of response of inverter-based systems – Paul Myrda

Inverter based systems have a different response than from synchronous generators and one that is potentially problematic for existing protection and control schemes.  Inverters are built to maintain very fast current control and so during the fault the current (typically, reactive current priority) output of the inverter is tightly controlled and maintained at or below rated full-load current.  If the fault is very close (i.e. terminal voltage of the inverter drops to say 5% or less) many will even block the inverter and so you’ll have zero current.  As such, if the inverter base generation is the only or major source in an area, protection can be challenging.  The major impact of inverter based systems on transmission system protection is their constant current behavior regardless of fault location.  Distance relays do not respond very well to that behavior so in cases where the sole source or majority of short circuit current is from Type IV WT or PV it can be a problem.  [VL30]

 

7.     Evolutions in AC Systems
 

7.1     Network Topology and Substation topology (Network Structure) to be assigned

Due to the economic pressure distribution system operators are forced to a cost-effective operation and maintenance strategy. In combination with the high volume of equipment simple and standardized tech solutions are necessary. Therefore the network topology tend to be radial-type or ring-type with opened sectioning points.

 

Figure xx [picture of AC/DC line from B5 Bologna paper]

 

 

Figure xy Meshed AC / DC network

 

 

[VL31]Figure zz  - ([09])

 

 

 

7.1.1     Substation Topology

 

7.2     Increment of power flow in a transmission corridor

 

As existing power lines reach maximum load ratings, options for increasing power flow in a transmission corridor will have to be explored as the siting of new transmission lines is expensive, time consuming, and in most cases, politically unpopular. 

The options include:

  • Raising the line voltage – typically an expensive proposition with limited results as the existing towers were not designed to operate at a higher voltage
  • Increase the conductor size/configuration – this option faces similar issues as the insulator strings are typically not designed to support the extra weight of heavier wire
  • In cases of double-circuit towers – one can operate the line as a 6-phase system (cf. §7.36.3 )
  • Undergrounding of new circuits using existing Right Of Way.
     

7.3     Six-phase systems – Mark / Jorge Cardenas / Alex

Multi-phase (employing phases more than three) power transmission systems have been investigated as a potential alternative to conventional three-phase systems for bulk power transmission at EHV and UHV levels in the past. The technology has been found to be effective to remove some of the limitations of three-phase system arising out of the use of extremely high voltages to increase power capability of lines, excessive rights-of-way (ROW) requirement and several environmental problems.

Six phase transmission provides a technique for reducing physical space requirements for transmission of Electrical Power, it provides a technique for increasing the power handling capacity of existing double circuit lines, reduce cost and minimize environmental effect with lower electrical fields.

Voltage operation level from individual phase-to-neutral is raised by over that for three-phase operation without changing the phase-to-phase voltage. In a typical double circuit line, this allows a transmitted power increase in the same ratio with no increase in losses (same currents). In other words, capability is increased by 1.732 times compared with a double circuit line working in a classical three-phase scheme.

7.3.1     Six Phase Fault Analysis

Fault impedances can be determined by measurements, matrix methods or Symmetrical Sequence Components.

7.3.2     Six Phase Symmetrical Components

The six-phase system has the characteristic operator . Figure 3.1 shows the six symmetrical systems of vectors. In addition to the positive, negative and zero sequence systems of vectors, there are two set of vectors which are repeating three-phase systems and one set which is repeating two-vector system.

  • Positive Sequence  (60º apart)
  • Second Sequence (120º apart)
  • Third Sequence (180º apart)
  • Fourth Sequence (240º apart). Opposite Second Sequence
  • Fifth Sequence (300º apart). Negative Sequence
  • Sixth Sequence (360º apart). Zero sequence

Figure 3.1 Symmetrical components systems for a six-phase system

7.3.3     Fault Combinations

Add short paragraph on the number of fault types – and a reference to the fact that current differential is a

[MGA32]

 

7.3.4     Challenges in the application of protective relays on a six phase system

Protection challenges in a six-phase system is primary focused on  HV line and transformer protection, starting from the assumption that the other components of the electrical network with remain three-phase, because there are no substantial benefits of using six-phase in generators and it is still early the development stage in distribution. In the other hand, for distribution, because the functionality required relays for three phase systems could be adapted with less difficulty.

a. Line Protection

  • Current Differential Protection: Pilot based current differential protection involves measurement of current on a phase by phase basis based on current differential principle. For that, the actual differential protection used for three phases could be used. In this case two sets are needed: one for phases a-c-e and the other one for phases b-d-f
  • Segregated Phase Comparison: Segregate phase comparison relays are needed. Majority of actual three-phase relays process composite current signals: e.g.
  • Distance Protection: Ideal a complete new Distance relay with six phase logic is needed. In the actual situation and for six-phase trip, two sets of distance protection could be used. One for phases a-c-e and the other for phases b-d-f.

b. Transformer Protection

Transformers provided in series with each set of three phase conductors of a double circuit line to achieve the six phase vectorial configurations need to be protected. Starting from the approach of the Figure 3.1, the protection system could be similar to the one used for three phase transformers. In this case, two sets of protection is needed.

Figure 7.8.1 Two 3Ph transformers inside the component to 3-to-6 Phase System

 

7.4     FACTS – Sankara

Flexible AC Transmission Systems (FACTS) are power electronic based devices and other static equipment that provide control of one or more AC transmission system parameters to enhance controllability and increase power transfer capability of power systems. FACTS devices are therefore being increasingly employed in power system networks to satisfy these requirements

In a power system, the transmission of power in a transmission line is mainly dependent on the sending and receiving end voltage levels, the transmission angle and the transmission line reactance.

To increase the power flow through a transmission system, one or more of the above parameters must be changed.  For example, the transmission angle can be increased with the use of a phase shifting transformer but such an item of plant is costly to purchase and install, and the transformer losses must be accounted for.  Increasing the transmission angle also pushes a power system closer to its stability limit, increasing the likelihood of system instability.  Also the transmission voltage level could be increased.  However, this would only be economically feasible if permitted by existing tower construction, and it would still be very costly to upgrade system insulation and switchgear.  Where such an approach is envisaged in the future, transmission lines could be constructed taking into account future operation at higher voltage levels.  Power flow could also be increased by reducing the inductive reactance of the transmission system by installing fixed series capacitors.  This was in the past found to be one of the most economical ways of increasing the power flow of the transmission system. [1][VL33]

FACTS devices can be broadly applied to increase the power flow or even to change the power flow by having a higher degree of control of the three key parameters of line impedance, phase angle, and voltage magnitude.  In addition, FACTS devices are used to increase the stability of the system and to regulate the system voltage.

 

There are two distinctly different approaches to realization of FACTS devices, the first is based on conventional thyristor technology and the second is by using voltage source convertors [1], [2], [4], [5].[VL34]

Different types of FACTS devices that are available are listed in the table below.

  • Thyristor Controlled Reactor (TCR)
  • Thyristor Switched Reactor (TSR)
  • Thyristor Controlled Capacitor (TCC)
  • Thyristor Switched Capacitor (TSC)

 

Table xx  Available FACTS devices

Conventional Thyristor technology based FACTS devices

Voltage source convertors based FACTS devices

Static shunt compensator of the following types:

thyristor controlled reactor (TCR)

thyristor switched reactor (TSR)

thyristor switched capacitor (TSC)

Fixed capacitor-thyristor controlled reactor (FC-TCR) thyristor switched capacitor & thyristor controlled reactor (TSC-TCR)

Static synchronous shunt compensator (STATCOM)

Static series compensators of the following types:

thyristor switched series capacitor (TSSC)

Fixed capacitor in parallel with thyristor controlled reactors (FC-TCR).

Static synchronous series compensator (SSSC)

Thyristor controlled phase shifters.

Unified power flow controllers

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure XY

  Conventional Thyristor based FACTS devices for different application [2]

 

 

 

7.5     [VL35] Stability

Transient Stability – move to issues[MGA37]

7.6     – how much extra weight / stress can existing towers support?

 

 

7.7     Interface requirements for Generators [VL38][VL39]

 

The increasing penetration of DG- units requires their active contribution to ancillary services. In this matter fault-ride-through and voltage support must be mentioned. Following the occurrence of disturbances in the transmission network the loss of DER-units results in loss of generation and of voltage support of the power system. Depending on the amount of lost generation the situation may worsen and in some cases lead to severe stability problems. As a consequence system operators now require the fault-ride-through capability of DER-units connected to the distribution grid. Furthermore system operators demand a reactive current from DER-units for fast voltage support during disturbances in the transmission network. In turn DG-units contribute to the short-circuit current while the network protection operates, also for disturbances in the distribution network.

 

Standards-based implementation – EN numbers – Mika

Several European standards concerning DER are established or published. These standards include :

EN 50438 : Microgenerators

  • EN 50549-1 and -2 : Requirements for generators connected Generators connecting to the LV/MV network
  • ENTSOE European Grid code : requirements for grid connection applicable to all generators.

 

 

Fig. 12. Reactive current from DER-units for fast voltage support during disturbances according to entso-e Working Draft Network Code, Europe

 

Fig. 1. Typical limit curve for FRT requirements (Nirmal – NZ)

The consideration of existing criteria gives a better understanding towards development of new criteria. FRT criterion has already been established and is mandatory in many countries. International experiences have also been considered while developing criteria for the New Zealand system and have been used as a comparison in Section IV[21][VL40]

[21] Tsili, M. and S. Papathanassiou, A review of grid code technical requirements for wind farms. Renewable Power Generation, IET, 2009. 3(3): p. 308-332.[n41]

 

Appendix 3[VL42]

7.8     Overload Management of lines - Jorge

7.8.1     Context

Growing demand for power is a major challenge for system operation worldwide, because utilities and operators are finding it near impossible to build new lines. Thus, transmission asset owners must explore the idea of increasing the capacity of the existing transmission lines. One way to do this is to maximize the use of the conductors on the towers, respecting the dynamic limits of the power system. To identify the optimum load for their HV and EHV lines, utilities and operators require a reliable dynamic rating system which shall be easy to install, so the line ampacity can be determined.

 

Overhead conductors are susceptible to ambient conditions such as air temperature, intense solar radiation, wind speed and wind direction, all of which can vary from one point along a line to the next. As such, the position of conductors can change, affecting a line’s vertical safety clearance and subsequently its thermal rating. Moreover, the factors affecting the thermal rating of a line are difficult to predict. Consequently, estimates to determine the maximum permissible current for transmission lines have been conservative (for instance, the Belgium NBN C34-100 standard). Power flows in many systems are becoming more widespread and unpredictable. Trading means that line loading has increased beyond expectations. In addition, due to environmental constraints, it can take between 4 and 10 years to receive permission to build a new line in many countries.

 

A Line Overload Management System (LOMS) [MGA43]can measure the sag of an overhead line in real time. The sag – resulting from load and ambient factors such as temperature, wind direction and wind speed – may be determined only by measuring conductor vibrations. Conductor vibrations are measured using accelerometers. Conductor sag shall be calculated based on these measurements using data processing (fast Fourier transform) and simple mathematical formulae. Once the sag is determined, a special software application can be used to calculate the line’s maximum permissible current and make appropriate forecasts.

 

A number of experimental LOMS modules have been fitted on the grids operated by two European Transmission System Operators (TSOs). Experience has clearly demonstrated that conductor sag is highly variable and heavily dependent on local weather conditions. The experimental LOMS real time monitoring system has shown to be accurate with a sag error margin of 2% which is consistent with a safe real time operation of the instrumented lines ([10]).

 

A LOMS module is shown in Figure 6.4-1:

 

Figure 6.4-1 – A LOMS module installed on a line conductor ([10])

7.8.2     Sag Check

Reference ([10]) shows the results obtained from field experience. Independent land surveyors measured sag at a given point over a period of 4 days between June and November 2009. Measurements were taken from 5 spans where the LOMS module was installed.

 

The LOMS module output data were sent to a TSO. Measurements showed a margin of error of around 20 cm, which was accurate enough to predict ampacity. Errors resulted from:

  • The land surveyor’s exact measurement time (margin estimated at 2 minutes);
  • The accuracy of the land surveyor’s measurement (margin estimated at 1%);
  • The LOMS module’s exact measurement time (margin estimated at 2 minutes);
  • The accuracy of the LOMS module’s measurement (margin estimated at 20 cm).

7.8.3     Calculation of the Ampacity

Real-time power line ampacity is calculated in three steps ([10]). The ruling span concept (CIGRE, 2007) is used to extrapolate the sag in one span to other spans:

  • Step 1 – evaluating present sag conditions (carried out regularly but not in real time);
  • Step 2 – real time effective weather conditions;
  • Step 3 – real time ampacity evaluation.

7.8.4     Impact of a LOMS

Some utilities and TSOs operate the EHV and HV grids based on permanent maximum permissible intensity levels calculated for homogenous areas according to climate and season ([10]). Based on conductor type, the region is split into different climate zones, each with one permanent permissible intensity level and two time-delayed permissible overload levels. Typically, for a given TSO, time delays are set at 20 minutes and 10 minutes. Seasonal operational procedures are based on a winter season, two intermediary seasons and a summer season. In addition to the monitoring carried out by the operator from the control centre, protective mechanisms set to the various permissible intensity levels (permanent and time-delayed) trigger safety measures when system transmission levels exceed the permissible maximum. These fully automatic monitoring and protection systems guarantee the safety of individuals and properties if any given stresses are not dealt with by the operator within the stipulated time frame.

 

To set up a monitoring system whose maximum permissible intensity data varies significantly in real time, automatic protection systems must currently be disconnected. Indeed, automatic protection system thresholds cannot handle the variable reference intensity. To disable the emergency protective mechanisms to the aid of the operator, operating regulations need to be modified and operators must have complete trust in the monitoring system reporting the permanent permissible intensity level and the temporary permissible overload levels.

 

For control centre operators to build their trust in monitoring systems that modify established practices, discussions must take place on the pros and cons of LOMSs and on which spans to equip. Research must be carried out on which critical spans to equip before installing and operating a system. The operator’s knowledge of sag, and hence the distance between the line and the ground, is of the utmost importance if the system is to be run safely and securely, especially in cases where only a few spans are equipped to provide an overview of the system’s operation.

 

Carrying out a test phase over several months without real operational use is also an important step for operators to take. It allows a mechanism’s performance to be monitored by comparing recorded results with topographical data. As indicated below, ensuring compliance with safety distances is critical in system operation. The test phase also helps to guarantee that data acquisition and data processing methods are reliable, since such data is recorded over a period of several months.

 

After the test phase, the system has to be put into actual operation. Operational procedures must be modified and the unique characteristics of real-time monitoring mechanisms taken into account. The procedures must also allow standard operation to be resumed without system monitoring in case the mechanism is not available.

 

The findings from the test phase confirmed that monitoring mechanisms were reliable, especially in terms of the physical measurements for line sag calculations. Furthermore, comparing the findings with measurements from topographical data helped to determine the level of accuracy of the entire data acquisition and processing chain. The results are accurate enough for industrial use.

 

The provided data revealed that, more than 99% of the time, system transit capacity was higher than the limit set using traditional methods, including both deterministic and probabilistic methods. Therefore, using the monitoring mechanisms for overhead lines should allow optimal system operation in real time.

 

However, forecasting and controlling stresses must be prioritized over pure real-time data processing. Indeed, having early access (a day or several hours in advance) to information on system transit capacity for the relevant period can optimize the impact of temporarily implemented measures (e.g. non-operation of the group or group operation at minimum power). Only using system transit capacity data in real time makes it difficult to adapt prevailing measures and control stresses.

 

So, to be able to make full use of the data provided by real-time monitoring systems, these systems must be combined with a predictive mechanism to provide a forecast of the allowable capacity within a given time period (2-3 hours or the day before). The predictive mechanism must provide a forecast – and a reliability rating based on the desired time-frame – using information provided by the monitoring system and local weather forecasts.

7.8.5     Brazilian Viewpoint

Reference ([11]) is the response to the following questions: “How much gain in transfer capacities has been obtained with the real time updating of line ratings? Are there examples of real time parameter updating (other than for line ratings) which also results in higher transfers?

 

There were no regulatory procedures in Brazil to establish line transmission current limits prior to 2005. The first initiative to deal with this problem was the Resolution no.191 published by the National Regulatory Agency for Electric Energy (ANEEL) on December 12, 2005.

 

The primary aim of this resolution was to establish a method to determine two capacities of a transmission line: the long term current capacity and the short term current capacity. In October 1992, CIGRÉ’s Working Group WG 22.12 developed the capacity calculation model for transmission lines voltage ratings 69 to 750 kV and ANEEL considered this to be the most suitable model to determine long term current capacity. Short term current limit is obtained by multiplying the long term current limit by a factor associated to the transmission line design temperature. The reason for this factor is that it becomes lower as the design temperature of the transmission line grows, as showed in Table 1:

 

Table 1

 

Design Temperature (Celsius Degrees)

 

50

55

60

64

65

70

75

80

90

Factor

1.42

1.33

1.26

1.24

1.23

1.19

1.17

1.15

1.12

 

The long term current capacity (LTCC) and the short term current capacity (STCC) are stated in the CPST Contract – Contract for Transmission Services, celebrated between Brazilian ISO and each transmission utility (ISO does not own transmission assets). It is important to emphasize that ANEEL’s Resolution 191/2005 also included a request that the ISO had to develop a methodology and the calculation of transmission line seasonal capacity. Thus, since 2005, the research on current capacity has continued and the ISO has been striving to implement this capacity.

 

The methodology to calculate the seasonal current capacity of a transmission line was developed by the ISO and considers four seasonal periods:

  • Summer Day;
  • Summer Night;
  • Winter Day;
  • Winter Night.

 

The seasonal current capacity of a transmission line is determined by a thermal risk, as described below:

  • Long Term Current Capacity (LTCC) is defined by the current level leading the conductor temperature to exceed the design temperature in no more than 15% of the meteorological conditions verified historically, as shown in Figure 6.4-2:

Figure 6.4-2 – Graph for Meteorological Conditions

 

Short Term Current Capacity (STCC) is defined by the long term current capacity multiplied by the factor shown in Table 1.

 

The thermal risk formulation is calculated by the following equation:

The methodology which defines the seasonal current capacity of a transmission line incorporates the following steps:

  • Step 1 – Based on geographic co-ordinates, the transmission line is divided in sections of approximately 10 km length.
  • Step 2 – For each section, meteorological data such as environment temperature, wind speed and solar radiation are simulated and factored over a period of eleven years. Thus, for each section, there are approx. 870,000 sets of data. These sets of environmental data comprise different origins and quality representing further difficulty when attempting to determine the ampacity of a transmission line. Therefore, a technical procedure called downscaling[1] has to be used to deal with such extra difficulty.
  • Step 3 – The final result is to obtain the shortest seasonal current capacity for each section of the transmission line in all the variation of its length.

 

The seasonal operational capacity of transmission lines will be stated in the operational instructions used by the ISO and transmission utilities.

 

Based on the above, the increase in LTCC and LTCC of the transmission lines established in the Resolution 191/2005 occurs in two stages:

  • In the first stage, there was an increase of these capacities based on applying the deterministic methodology developed by CIGRÉ. This served to standardize the method of calculation among all agents of transmission and review all CPST contracts.
  • In the second stage, additional increase of LTCC and STCC will be based on the application of seasonal capacity considering the Probabilistic Risk. This second stage is being implemented by the ISO, therefore there are still no results verified in real-time operation of the energetic gains that can be obtained. However, off-line studies show that the average gain in the capacity of transmission lines when applying the method of seasonal capacity is about 19% higher compared to values obtained by applying the deterministic methodology currently used. Figure 6.4-3 depicts an example:

VistaDeCima  LT 500 kV SGP-OP.jpg

Figure 6.4-3 – Graph for Meteorological Conditions

The major challenge on which the ISO and ANEEL are working is to establish a standard for monitoring of transmission lines. Currently, there are 15 meteorological stations installed along two transmission lines in southern and southeastern Brazil, for which it is possible to calculate the value of real-time long term current capacity. The ISO will seek to increase the number of meteorological stations connected to other parts of the transmission system in order to obtain the dynamic ampacity of the transmission lines in a considerable area of the Brazilian Interconnected Power System.

 

7.9     Networks with increased Cable Lengths

7.9.1     Distribution grids– Mika (contributions from Lars)

Undergroud cables are a major part of distribution network in the city and urban areas. In some cities almost 100 % of the distribution network is already cabled, many cities have mixed network with underground cables in the city center and overhead lines in suburban areas ([14]).

 

Protection in the cabled distribution medium voltage network differs from one in overhead line networks. In pure cabled network autoreclosures are not often used because there are not self-extinguishing faults like in the overhead line network. There are not usually high impedance earth faults either in the cabled network. The shield of the cable can usually carry around 50 % of the earth faul current. Sensitive high impedance earth fault protection is therefore not necessarily needed.

 

If the earthing method of the underground network is appropriate (resonant grounding system or ungrounded system with low capacitive earth fault current level) and earthing conditions are good there is an option to use earth fault protection in alarming mode without tripping. This leads to lower amount of customer outages. In the overhead lines all the earth faults have to be tripped because of dangerous fault arcs and earth voltages. When the earth fault is not tripped fast, higher healthy phase voltages during an earth fault (up to phase to phase voltage level) cause more sress to the other network and can more easily evolve to cross country faults. Cross-country faults have to be tripped with overcurrent protection or special protection stage for cross-country faults. The amount of cross-country faults depends on the type, condition, structure and age of the underground cable network.

 

In the cabled network and especially with resonant grounding system intermittent earth faults are quite common. Low earth fault current level enables the fault current extinguish at the zero point and ingnite again at the peaks of instantaneous voltage. Special functions to detect these intermittent earth faults have been developed and used already for some years. These functions can be located at the bay level IED’s or at IED’s specialized for resonant grounding system fault detection.

 

Automatic fault location has been quite widely used new feature in medium voltage networks for some time. There are two main methods for that:

 

  • centralized fault location, functions located at the substation level and at the Distribution Management System (DMS)
  • distributed fault location, functions located at the secondary substations (fault current passaging information)

 

For short circuit location the centralized system works well with both underground and overhead line networks. IED (overcurrent function) in the substation can measure the fault current and send it to DMS system, which can show the exact location of the fault graphically, when comparing the measured current with calculated short circuit values. If the earthing method of the medium voltage network is solidly or low impedance grounded, all the fault currents are almost like short circuit currents. For other earthing methods the earth fault location is more complicated. Load currents and inaccuracies in the zero-sequence impedance cause errors to the earth fault location values. For zero-sequence impedance the teoretical zero-sequence impedance can differ a lot from the real one. With overhead line network the error is not as large as with underground cables. This method has been mainly tested with overhead network. For cabled network all the medium voltage line impedances should always be measured also after changes in the network structure.

 

The distributed fault location is more used in the underground cable network. During fault situation there is only need to isolate fast the faulty part of the network in open loop structure. In distributed fault location system short circuit and earth fault indicators located into transformer stations are used to send an automatic fault passaging alarm to SCADA and DMS –systems. With these fault indications the faulty part of the network can be located and isolated more quickly with manual or remote controlled switching operations in the network.

 

7.9.2     Transmission networks – Volker

Mainly due to environmental considerations and to acceptance problems, it has become more and more difficult to build high voltage overhead lines. One solution for TSO consists in constructing very long AC underground cables, even atr higher voltage levels. For this reason the number of projects involving

long underground lines has increased significantly over last years and is expected to increase further.

Several projects of long AC underground lines have been reported (e.g 225 kV 65 km cable in the South East of France [reference], including AC connections of offshore wind farms to the onshore grid. In general, these cables require shunt reactors at the line ends in order to compensate their capativie current.

 

These unusual cable lengths give raise to many studies which have to be conducted to verify the cable impact on the grid and on equipments. These studies include:

  • Frequency scan studies to detect potential dangerous resonance with the network ([13])
  • Time domain studies to define the requirements for high voltage equipment ([13]) and to determine the adequate protection system.

 

The important length of this cable requires special attention on protections relays, particularly on differential protection. Because of its behavior, the line differential protection is at risk of malfunction (loss of sensitivity or untimely operation).

([12]) studies the operation of differential protection on a structure of this type. It was verified that protections have a sufficient sensitivity and remain stable despite the many transients during operation of HV components. A limit on the operation was however indentified for the short-circuit power level.

Since the faults that may appear on the link are mainly single-phase faults, the setting of a distance protection will require an exact knowledge of zero-sequence characteristics of the cable. Before the commissioning of the cable, it is recommended to make measurements of its electrical characteristics.

It has been estzablished that a method to set distance protection on long cables should be defined.

In some cases, significant overvoltage on the cable can occur when it is operated as antenna. If necessary, it may be necessary to use intertrip schemes destined for the circuit breaker in the remote substation when the cable circuit breaker opens at the other end.

 

Another particularity of the protection of large underground lines are the compensation reactances. Transient studies show that in some cases a separate energisation of reactances and cables is required in order to avoid high transient overvoltages. This may lead to a varity of transient topologies for the cables and the associated equipments, which have to be all taken into account correctly by the protection system. Obviously, the reactances have their own protections sytem against internal faults.

 

 

 

 

7.9.3     Automation in distribution networks - Mika

Traditionally the medium voltage network and its switching devices and transformer stations have been manually operated. Investing into the MV network has been very economic driven in the past decades and cheap and simple solutions were preferable. Lack of technically and economically feasible communication channels has also been an obstacle for distribution network automation in the past years. 

 

The need to shorten the customer average outage times (SAIDI) and individual fault outage times (CAIDI) further leads to the need to implement more MV network automation systems. Another stimulating factor has been the fast improvement of communication technology (mobile radio communication, optical fibres...) and lowered prices of communication. Faults in the MV level have the largest contribution into the outage levels of Distribution System Operator utilities. The MV network automation produces in fact one of the biggest impacts on the future network from DSO’s point of view. Ongoing and future automation implementations in the MV -level can be for example:

 

  • Automation of secondary MV/LV transformer stations
  • Automated MV level switching procedures (self healing network)
  • Moving towards closed loop MV network
  • Electricity storage systems
  • Microgrids

 

 

Picture X. Different kind of choices for medium voltage customer connection and improving quality of supply in Helsinki.

 

7.9.3.1    Automation of MV/LV transformer stations

 

DSO’s operating in rural areas have in some countries already during decades invested into remote controlled switching stations located into some strategic locations. Large amount of faults caused by climate conditions and long distances for operating personnel have been arguments for these investments which help to shorten the outage times.

 

Nowadays and in the future the rural DSO’s will invest not only for remote controlled switching stations but also for:

 

  • Circuit breaker stations with local relay protection, usually overcurrent and earth fault functions used. CB stations are located at the transition points of cable and overhead networks and at the most considerable node-stations of the MV network.
  • Remote controlled disconnector stations with help of fault location functions and other complimentary monitoring information. Centralized substation IED level fault location with help of DMS -system is mostly used to help to locate the most probable point of fault. Electricity can be restored fast to the healthy parts of the network with remote controlled disconnectors without any test switchings. In radial network the exact fault place has to be found fast to enable the fault repairing work and restoring electricity to the rest of the network as rapidly as possible.

 

In the city networks and cabled networks DSO’s have also started and will start to implement network automation. Often in the city area the MV switchgears are so called simple ring main units, RMU’s. Remote controlled disconnectors at transformer stations, fault location information and transformer station monitoring will be the main features of distribution network automation in the city networks.

 

Remote control is most commonly used in normal service operations, but the biggest advantage can be acheved in the fault cases. In the city areas the heavy traffic and admittance problems to transformer stations located into non utility owned facilities are issues which slow down the access time to do manual switching operations in fault cases. As the most important feature can be seen isolating the faulty MV network part via remote switches fast with help of short circuit and eart fault indicators located into transformer stations. If there already are instrument transformers or novel measuring devices installed at the transformer station, installing of the fault locators will be easier and more cost effective. If there are not measuring devices, most reasonable and cost effective way to realise the fault locating should be determined. Short circuit location is easy with magnetic field detection based indicators, but earth fault indication can be more difficult especially in cabled networks. Io-measurement and often also Uo-measurements are needed to enable fault current detection.

 

When planning the locations for the secondary substation automation, there have to be calculations which take into account the costs and benefits. Usually it is reasonable to locate the automation devices to feeders and locations which feed the most important part and largest amount of customers and which have the highest customer interruption costs. The open loop points of the distribution feeders and transformer stations in the middle of feeders are often the most economical locations for automation and remote control equipment. When thinking the location of the faulted part of the network as exactly as possible also the rest of the secondary substations should be minimumly equipped with fault locators which send their information to SCADA and DMS.  

 

 

Picture x. Location of the secondary substations equipped with remote control.

 

 

Monitoring of the secondary substation at MV and LV level is one advantage when thinking distribution network asset management and power quality control. Overload monitoring of the distribution transformer and low voltage power quality measurings are examples of that.

Picture x. Example of secondary substation automation system in Helsinki.

 

 

7.9.3.2    Automated MV level switching procedures (self healing network)

  •  
  • One step forward is to automate part or all the switching operations of automated secondary substations. There are at least two options where to place the central intelligence of automation. It can be placed into substation level automation. It is OK if the substation communication network is enlargened outside the substation and the substation is the only one which takes responsibility for feeding the automated transformer stations. The problem is that usually there are several substations which can feed one secondary substation. This is reality at least in the city networks. Therefore centralised automating model is better at least for city circumstances. Utility’s SCADA-system or Distribution Management System are platforms for these automated distribution fault isolating sequences, called also as self healing network. For example when this automating sequence system is located into DMS system, the intelligence should need the on-line information from:

 

  • substation CB-positions
  • substation SC and EF detection information
  • substation SC or EF location information, if used
  • secondary substation switching device positions
  • secondary substation SC and EF -detection information and/or measured secondary substation fault currents for further conclusions

 

Problem with fully automated fault isolation system is that the fault location information is not always reliable whereby false operations are possible. First step or alternative for fully automated system is fault isolating sequence which only advises the switching sequence, all real switchings has to be still approved by human operator. There are for sure some pilot systems or limited fully automated MV level self healing network applications in use, but larger and DSO-wide systems are just on pilot or research stage.

 

7.9.3.3    Moving towards closed loop MV network

Next step further from open loop MV solution is to use closed loop solution. This will be most feasible in cabled city networks, which already have open loop solutions. Closed loop solution means moving towards technology already used in the substations and protection technology used in meshed transmission networks. Operating and protection of closed loop MV solution is not any more possible with simplified RMU disconnector transformer stations. Remote controlled devices, circuit breakers, instrument transformers and protection relays are needed for the secondary substations, desing of them goes towards primary substations.

 

Protection of closed loop solution is not any more possible only with overcurrent and earth fault protection, because load and fault currents can flow in any direction. Differential protection could be most suitable main protection solution for cabled networks, for overhead line network distance protection can also be used. Earth fault protection depends on the earthing method, but with resonant grounded and isolated networks separate EF protection is needed.  In most cases the sensitivity of differential and distance protection functions is not sufficient. If differential protection function is used for short circuit protection, then current comparison function could be applied for earth fault protection. For back up protection traditional directional SC and EF protections are suitable in most cases. Busbar protection function can also be added to secondary substations, with dedicated busbar protection unit or using the feeder IED’s. These substations need also modern communication channels for remote connection and also for protection communication. IEC 61850 standard station level communication can be used. For larger amount of secondary substations there can be a common gateway for remote control direction. 

 

 

 

 

 

Looped feeder overview_org pic_ppt modified

 

Picture X. One solution from Helsinki for closed MV loop main protection.

 

 

Picture x, Communication links for control and protection of MV closed loop solution, Helsinki.

7.9.3.4    Electricity Storage systems

 

There are already some commercial MW –level distributed electricity storage systems available. The system consists of MV (or LV level) equiment, step up or isolation transformer, DC/AC inverters and batteries. These systems can be used for :

 

  • UPS-solutions,
  • reactive power compensation,
  • electricity peak shaving,
  • frequency regulating (TSO level)

 

Picture x. Example of modular container electricity storage solution, 1-4 *0,5 MW, max. 2 MW, example : 1 MW, 1 MVAR, 0,5 MWh.

8.     Impact of Technology Trends on Protection and Automation of Primary and Secondary systems

 

19-10-2012 [VL44]

Evolutions in HV equipment may have a huge impact on the protection- and control system. One of the most impact evolutions is the use of Non-Conventional Instrument Transformers (NCIT) or the use of merging units connected to conventional instrument transformers, because this enables important modifications and simplifications in the substation architecture. The main reason for this gain of flexibility is the use of fiber optics for the transmisision of analog values. The distance between instrument transformer and control or protection device is thus no longer constraint by the maximal length of wired secondary current- and voltage circuits.

The possibility of interfacing digital inputs and outputs (eg. Trip command) of HV equipment with the process bus has also an impact on the protection- and control system.

If HV equipment based on a completely new technology is introduced, the corresponding control- and protection system will have to be designed. This is the already the case for HVDC links and may soon be the case for fault current limiters, static circuit breakers or transformers or supra-conducting feeders.

It also has to be metionned that the evolution of technology will also have an impact on the control- and protection system itself. This concerns the communication systems, but also the hardware developpements which enable a higher level of integration of the control system.

This section discusses some of these issues.

 

 

16/05/2012[VL45]

 

Substations provide all means (busbars, circuit breakers, isolators, and grounding switches) to switch-on, re-direct or switch-off the power flow for operational reasons or in case of emergency. Indications about foreseen trends in substation technology are given in Table 3:

 

Issues

Now

Future (next decade)

Use of non-conventional instrument transformers

Pilot projects with proprietary connections

Commodity for new substations and for replacements including standardized serial FO links

Interface for conventional instrument transformers

Hard-wired

Serial FO links introduced in case of refurbishment

Switchgear interface

Hard-wired

Serial FO links introduced in case of refurbishment

Interfaces for needed condition based monitoring

Partial

Comprehensive

Integration of IEDs into switchyard devices

Separated IEDs

Integrated IEDs

Table 3 – Key issues in the evolution of switchyard technology

 

Issues

Now

Future (next decade)

Processing capability

Acceptable computation power
(≤ 1 GHz processor rate)

More powerful computation

(> 1 GHz processor rate) & Multicore & Multiprocessor

Hardware architecture

Application function processing integrated with sensor and actuator interface in one dedicated IED

Split between (industrial) application function processing platform and dedicated interface IEDs

Communication

(see also IEC 61850)

Dominated by copper wires;

no standardized dual port redundancy

Dominated by serial fiber optic links in case of  new systems and retrofit;

dual port redundancy

Protection

Preconfigured protection

Fault evaluation centralized

Conventional algorithms

Adaptive protection

Embedded fault evaluation

Selective backup protection

Wide area protection (SIPS)

Advanced algorithms driven by NCIT

Monitoring

Limited monitoring of

  •     power system
  •     automation system

Comprehensive monitoring of

  •     power system
  •     automation system

Control

Manual control by operators

Operator control strongly supported by automatisms

Function integration

Low: many dedicated IEDs

High: optimal, least acceptable low number of IEDs

System engineeringSystem engineering

Variety of disconnected proprietary tools for layout design, simulation, specification of requirements, programming of controllers and design of visualization. Basic functionality of IEC 61850 tools

Advanced IEC 61850 tools integrating all stages of system design into an integrated process. Model-based software engineering based on open standards, connecting all stages from single-line diagram drawing to executable code generation, simulation and  deployment, free allocation of functions across computational resources, dependable communication, automatic balancing of resource consumption and system performance[VL46]. Advanced IEC 61850 tools considering e.g. free allocation of functions, resource consumption and system performance as well as synergy of Automation lay out and communication

System maintenance

Individual maintenance procedures

Common maintenance procedure by SCD file and version handling according to IEC 61850 

Power system integration

Dedicated task splitting and overhead by duplication

Intelligent task sharing

IEC 61850

Successfully introduced

Request for extensions

Comprehensive coverage of power system domain (seamless comm.)

Table 4 – Key issues in the evolution of substation automation technology

 

 

8.1     Impact of HVDC on AC protection - Volker

 

The Technical Brochure from CIGRE JWG B5/B4.25 ([15]) adresses a list of phenomena related to LCC DC links which interfere with the protections of the AC network. The identified issues are:

  • Impact of harmonics created by converter poles on line- and transformer protections.
  • DC ground loop currents may cause saturation in power transformers or current transformers with adverse effects on AC protection.
  • Discharge current of DC cables after interruption of DC operation may cause saturation in power transformers or current transformers with adverse effects on AC protection.
  • Limitation of fault current infeed by converter poles. This may lead to situations where AC protection do not pick up or do not trip due to low fault current.
  • AC voltage distortion as consequence of non-symmetric DC operation of the link may have adverse effects on AC protection.
  • Fast adjustment of converter parameters in case of AC or DC fault. This may cause transients leading to untimely trips or erroneous fault direction indication by AC protections.
  • DC pole commutation failures may cause transients leading to untimely trips or erroneous fault direction indication by AC protections.

The TB ([15]) describes these effects and indicates possibilities to mitigate them. In some cases, the mitigation is not limited to the settings or parameters of the AC protections, but includes modifications in the design of the AC protections and / or of the HV part of the substations. Straightforward mitigation strategies are identified for each issue and a general incompatibility or deterioration of AC protection performance associated to DC networks is not anticipated. Comparable issues can be expected for DC links using VSC technology. The issues identified above remain to be investigated in detail for VSC.

 

8.2     Impact on SAS Architecture - Volker

 

8.3     Impact on network and substation automation - Volker

In addition to the power transmission, HVDC stations can also be used as FACTS. This is especially true for HVDC stations using VSC technology, which allows control of the reactive power injected or absorbed by the converter pole almost independently of the active power throughput. HVDC links of course also allow to control the active power flow.

These features need to be controlled and automated. The converter poles normally have interfaces with the DSAS and the protections of the substation and with the network control centres. The latter interfaces are used to choose among a predefined list of algorithms the operation mode of the HVDC converters, including active- and reactive power injections. In case of a network incident (on the DC link or on a neighbour AC line), the behaviour of the HVDC link must be consistent and coordinated with the rest of the network. It also must be compatible with AC protections (cf. previous paragraph), Special Protection Schemes (SPS) or Wide Area Protection Schemes (WAPS) and regulator characteristics [MGA47]of electrically close generation plants. In most cases, HVDC links will require to modify existing or to create new SPS or WAPS. These features may require the development of control algorithms implemented in the local DSAS and / or distributed over several substations and control centres. Given the different network topologies and constraints, such algorithms would require a special definition and specification for each case.

Usually HVDC poles are sensitive to tripping of their AC connection. The control of the converter poles usually receive CB trip commands in order to prepare the valves for the imminent opening of the AC circuit breakers in order to avoid voltage- or current stress for the valves of the converter poles. In given network topologies, the opening of other circuit breakers than those of the HVDC AC feeders may cause the converter poles to de-energise. This is for example the case if the busbar section connected to the HVDC AC feeder of a converter pole is connected to only one incoming AC line and the distant circuit breaker opens. DSAS or area control algorithms may be able to provide a correct trip information for the converter control system even in these exceptional cases.

The sensitivity of the HVDC converter poles of different technologies should be determined and mitigation approaches have to be identified, since a 100% reliable information of the converter control system of imminent tripping is not feasible.

 

8.4     Protection of HVDC networks - Volker

Most HVDC installations at this date are either back-to-back converters or associated to two terminal HVDC lines. The need of multi-terminal HVDC lines or even HVDC networks has been identified as a response to the challenges of long distance bulk transmission (like the Medgrid or Dessertec projects) and integration of high-sea windfarmes (e.g. the North Sea Super Grid – NSSG [2]).

In point-to-point or multi-terminal HVDC links, a failure of the transmission line requires a shut down of the converter poles and of the power transmission. On the contrary, in a HVDC network, a failure of a transmission line must not lead to a shut down of the complete network. For this reason, selective fault detection and –elimination, like in conventional AC networks, is required. Since neither AC protections nor AC HV equipment (measurement transformers, circuit breakers) is compatible with HVDC operation, these equipments and the associated concepts have to be developed for DC operation. The fact that converter pole operation under faulted conditions leads to a very fast shutdown (in some microseconds or faster) requires new and efficient concepts both for the selective fault detection and for the circuit breakers.

As far as protections are concerned, impedance measurement under steady state conditions is difficult in DC and would lead to an unacceptable tripping time. Also, a fault in a DC line or cable first discharges the phase-to-ground capacity of the line. Therefore, ongoing investigation is focussed on transient algorithms for fault detection and –location. The possible rapid change of state of the adjacent converter poles has to be taken into account.             
Concerning circuit breakers, AC equipment takes advantage of the passage of zero of the current in each period for opening the circuit breaker poles. Since this passage does, by definition, not happen in DC, other principle enabling opening of circuit breakers on DC lines have to be developed. If those approaches include the injection of transient currents (e.g. in order to obtain an artificial zero crossing of the current), the protection algorithms have to take these phenomena into account.

Investigations and feasibility studies have been initiated by several organisations [3] and CIGRE JWG B4.B5.59 has been launched to cover the issue of protection of these networks.

 

8.5     Protection of local LV DC networks

In addition to bulk HVDC transmission networks, the appearance of local LV DC networks (e.g. limited to an area or even a building) is possible. As a matter of fact, most computer equipment internally require LV DC supply, and AC/ DC conversion for each equipment is costly both in energy losses and in hardware. Today, this type of network has a radial structure and is quite simple to protect. If, for reliability reasons, meshed DC networks are installed locally, similar issues for selective fault detection and-elimination arise as for the bulk HVDC networks discussed above. Owing to lower voltage and currents involved, LV DC-fit circuit breakers and protections may be easier to develop for these networks.

 

8.6     Impact of FACTS on protection and automation – Sankara

The impacts of the Flexible AC Transmission Systems (FACTS) Devices (which includes all types of FACTS devices including the fixed series capacitors, thyristor controlled (series / shunt) capacitors and reactors such as TCSC / SVC, Converter type devices such as  STATCOM / SSSC influencing the operational features of the distance protection relay have been discussed in many literatures.

The effect of different types of FACT devices on distance relays has been studied in the literature and is summarised below:

  • Fixed series capacitors
  • STATCOM
  • SSSC
  • Conventional thyristor controlled FACTS devices type TCSC.

 

In this guide we focus on converter-based FACTS devices influencing the performance of the distance relay under various system conditions. 

The two unique operating features of converter-based FACTS devices in addition to their capability to internally generate reactive power, have so far been established are:

  • The capability to maintain maximum compensating current / voltage in the face of decreasing line voltage / current, which results in superior characteristics for shunt / series compensation.
  • The ability to exchange real as well as reactive power with AC system and thereby provide independent control of real and reactive power flow in the transmission system.

 

STATCOM by design can feed inductive or capacitive current even at very low line voltages, rather independently of the line voltages.  This enables the STATCOM to emulate a capacitive reactance or inductive reactance at the point of connection.  This is a most important point from the distance relay point of view under different system conditions.

 

 

8.6.1     Reach of Distance relay

The above scenario, where the apparent impedance calculated by the distance relay is higher than the actual fault impedance is termed ‘under–reaching’ of distance relays, which can happen in strong systems. In a weak system the distance relay will ‘over-reach’, where the apparent impedance calculated by the distance relay is less than the actual fault impedance. It has been reported that the fault location plays an important role.  The ratio of the compensating current provided by the STATCOM to the relay current will increase as the distance to fault location increases. Hence the errors in the impedance measured by the distance relay increases with the increase in the distance to the fault. Accordingly, the under reaching or over-reaching effects have been found to increase beyond 75% of the transmission line length. Simulations has shown that for a normal zone-1 setting of 80% of the line length, with 20% allowed  to account for the CT/ CVT errors relay  mal-operation has occurred at the boundary region of the zone-1.

With the increase in ratio of shunt compensating current to the relay current (i.e.) when the ratio of the compensating impedance to the impedance from the relay location up to fault is lower, the fault locator accuracy of the distance relay will also vary, giving wrong fault location values.

 

8.6.2     Effect of Load Angle

As the load angle increases the impedance calculated by the distance relay also increases. Published results has indicated that the distance protection under-reached for a fault at the reach point as the line loading increases. It is thought that this is due to the STATCOM’s attempts to maintain the voltage at the mid point to its nominal voltage, i.e., the voltage at the midpoint dips from its nominal voltage and this requires the STATCOM to produce more reactive current to boost the voltage at the mid point and thus increases the apparent impedance seen by the distance relay.

 

8.6.3     Relay Operating time

The operating times for faults at 75% of the line length has been found to be higher in systems operated with a STATCOM installed at the mid-point of the transmission line than the operating times in systems with out a STATCOM installation .

 

8.6.4     Unsymmetrical Faults

Even during unsymmetrical faults, the STATCOM provides equal compensation for all the 3 phases. This would mean increased voltages in healthy phases. This result in unequal compensation of all the 3phases which would in turn results in reduced compensation as the equivalent voltage will not be a true representation of the faulted phase.

Therefore during unsymmetrical faults the magnitude of STATCOM compensation would depend on factors such as pre fault load and hence the load angle, phases involved in the fault, system strength and fault location. This effect can result in incorrect impedance measurements which have been highlighted by simulation results.

The overcompensation on the healthy phase during un-symmetrical fault condition shall result in increased reactive current in healthy phases. This increases the possibility of wrong phase selection particularly for the relays using the current based phase selection. This point was confirmed during the testing of a commercial relay in RTDS.

 

8.6.5     Symmetrical Faults

It has also been found that for a 3 phase fault at a particular location the measured impedance can tend to infinity. This happens when the compensating impedance is equal to the line impedance between the shunt FACTS device and the fault. Other relays in the system see this condition as a power swing.

 

23/10/2012[VL48]

 

Centrally controlled fault location and service restoration may be inappropriatly time consuming, and unreliable given the possibility of communication failures in case of major power supply disruptions. Distributed control of these SmartGrid functionalities can help in overcoming these issues. It has been demonstrated that FLISR is achievable with fully distributed on account of using IEC 61499 distributed architecture and multi-agent control ([16]) ([17]).This solution promises higher robustness of power distribution systems, shorter power restoration and more efficient design of automation software.

 

8.7     [VL49]Effect of Power Electronic Conversion on Protection

 

8.7.1     Transmission and Sub-transmission – Nabil + Volker

 

8.7.2     Distribution – Nabil

 

The interconnection of Distributed Generation (DG) brings a great change to the configuration of the utility distribution network. It is expected that most DG will be installed in rural networks, however, future Smart Buildings may be able to inject load into the network.

 

Nowadays, the most common configuration in distribution systems is radial. In this type of configurations only one source feeds a downstream network ([18]). With the connection of DG, in case of fault, the system can lose the radial configuration, since the DG sources contribute to the fault and therefore, the system coordination could be lost.

The connection of new generation sources in the distribution system modifies the power flow, customer’s voltage conditions and the requirements of the utility equipment. In a fault situation, distributed generators modify the current contribution to fault, and therefore it influences in the behaviour of network protection. The influence will depend on the number, type, location and size of DG ([19]).  Thus, the characteristics of power equipment and the coordination system, which were established without considering the contribution of distributed generation, must be checked when DG is going to be connected.

One of the main problems in system protection when installing distributed generation is the ability of protective relays to detect external faults, especially in asynchronous generators and in generators connected through converters. This is the case of wind turbines. Fig 2 below shows the most installed configurations of wind turbines.

Fig2 Wind turbine configuration

 

Fig2a shows the fixed-speed wind turbine with asynchronous squirrel cage induction generator (SCIG) directly connected to the grid via transformer.

Fig 2b represents the limited variable speed wind turbine with a wound rotor induction generator and partial scale frequency converter on the rotor circuit known as doubly fed induction generator (DFIG).

Fig 2c shows the full variable speed wind turbine, with the generator connected to the grid through a full-scale frequency converter.

- Wind turbines before short circuit contribution

Wind turbines contribution to short circuit is analyzed in this section. Fig 3 shows the current contribution of an SCIG generator.

 

Fig 3 Asynchronous generator rms current contribution

 

Unlike synchronous generators, induction generators do not have field windings to develop the required electro-magnetic field in the air gap of the machine, so induction generators can not work without external power supply. Therefore, under fault conditions their air gap flux drops quickly and their contribution to the fault is usually negligible after two or three cycles. During the first cycles the contribution of the asynchronous machine is not despicable, therefore, neglecting the induction initial short circuit current could lead to errors in the choice of the protective relaying, switching equipment and phase settings of the protective relaying ([20]).

In order to compare, Fig. 4Fig. 3 shows a synchronous generator short circuit current contribution.

 

Fig. 43. Synchronous generator rms current contribution

 

Comparing both figures, it is observed than the main difference between the short circuit current contributions of a synchronous generator (Fig. 4Fig. 3) and an asynchronous one (Fig. 3Fig. 2) is the speed with which it drops. Then, the detection of the fault by the protective device must be done before the current contribution drops. The response time of the overcurrent relays is around 40 to 60 ms, therefore, if the relay settings for the instantaneous pick up current do not take into account the response time of the relay and the time in which the contribution drops, the protection could not detect the fault.

 

DFIG stator is connected directly to the network while its rotor is connected to the network by means of a power converter which performs the active and reactive power control. A voltage dip causes large currents in the rotor of the DFIG to which the power electronic converter is connected, so a high rotor voltage will be needed to control the rotor current. When this required voltage exceeds the maximum voltage of the converter, it is not possible any longer to control the desired current ([21]). This implies that a large current can flow, which could destroy the converter.

 

In order to avoid breakdown of the converter switches, a crowbar is connected to the rotor circuit. When the rotor currents become too high, the converter is disconnected and the high currents do not flow through the converter but rather into the crowbar resistances. Then the generator operates as an induction machine with a high rotor resistance. When the dip lasts longer than a few hundreds of milliseconds (Tmax_crowbar), the wind turbine can even support the grid during the dip (Refer to Section of Ride-through faults WT).

 

This behaviour is shown in Fig 5 a 200 ms short circuit has been simulated. The bold line shows the crowbar state, at t = 0.5 s, the fault is produced and the crowbar is activated. After Tmax_crowbar ms, the crowbar is deactivated; the rotor is connected to the converter and the control of the converter limits the current contribution to its nominal values.

Fig. 54. DFIG rms short circuit contribution

 

Full converter contribution depends on the converter behaviour before short circuit. Most converters can not supply important currents under external fault conditions; usually no more than 1.2 to 1.5 times their rated current of the converter. For that reason, it can be modeled as a synchronous generator with the option “Limit Maximum Phase Current” activated. In this case, fault detection schemes using overcurrent principles that are universally applied, are not usually effective. Wind turbines units that use this technology must rely on other methods such as abnormal voltage or frequency sensing to detect faults on the area power system.

- Wind turbine overcurrent protection analysis

In order to analyse the protection behaviour before wind turbine contribution, and the ability of protection software to model this response, a given network has been modelled in PSCAD/EMTDC

The wind turbine model in PSCAD/EMTDC consists of the asynchronous generator model, the drive train model and the rotor model. The generator and the drive train models have been developed by using the models available in the PSCAD/EMTDC library.

Shaft system influence must not be neglected since the shaft oscillations result in fluctuations of the voltage, machine current, rotor speed and other electrical and mechanical parameters. The shaft system has been modelled by using the PSCAD/EMTDC library multi-mass model. The literature about modelling indicates that a two mass model is adequate to model the drive train in wind turbines.

The rotor is a complex aerodynamic system that can be modelled with different detail levels. When electrical behaviour is the main point of interest, an algebraic relation between wind speed and mechanical power extracted is assumed to model the rotor behaviour, which is described by the following equation ([20]) :

(1)

Where P is the power extracted from the wind [W], ρ the air density [kg/m3]; A the swept area [m2] and Vw the wind speed [m/s]. Cp is the performance coefficient that depends on λ, the tip speed ratio:

(2)[MGA50]

Fig. 6Fig. 5 shows the modeled network and the faults (A, B, C, D and E) simulated in PSCAD/EMTDC. The red squares represent the points in which the overcurrent protections are connected.

Fig. 65. Modelled network

 

Fig. 7Fig. 6 shows test A results obtained with PSCAD/EMTDC. The value of the maximum current contribution from the wind farm is 109.4 A.

Fig. 76. CAPE and PSCAD results for test A

 

Table 3Table 1 shows the tripping time obtained by PSCAD/EMTDC. Breaker opening time (60 ms) has been taken into account in these results. In test A, the inverse time overcurrent protection (51) detects pickup, but, as it has been seen in the previous sections and in Fig. 8, when the trip signal is active (0.402-0.06), the short circuit current contribution is almost zero and the real protection would not trip.

Table 31. Overcurrent times by CAPE and PSCAD/EMTDC

Test

Breaker

Protection function

Operation time

A

BRK_AT

Inverse time

No trip

B

BRK_C1

Instantaneous

0,0988

C

BRK_C1

Instantaneous

0,1013

D

BRK_C2

Instantaneous

0,0994

E

BRK_C2

Instantaneous

0,1050

 

For protection system analysis, PSCAD/EMTDC recommends the synchronous generator model with the option “Limit Maximum Phase Current” activated to model the wind turbine behaviour. This model can be useful to model the full converted wind turbines (Fig 2c), since the converter limits the short circuit current to 1.2 or 1.3 times its rated current value. Nevertheless, the DFIG and SCIG short circuit behaviour do not show a good agreement with the shown by this model.

Induction machine short circuit contribution drops after a few cycles, but it is not negligible in this few first cycles. Due to it, wind farm contribution must be taken into account to set the instantaneous overcurrent protection, but not for the inverse time overcurrent protection.

If the asynchronous generator current contribution is neglected, the errors in the instantaneous protection setting can be significant in those zones with high wind power penetration. In case of modelling the asynchronous wind turbine as synchronous generator with a correct value for transitory impedance, like in this analysis, the initial behaviour (first cycles after the fault) is correct, but in few cycles the contribution of the asynchronous generators drops to zero and after those cycles the representation by means of this model no longer is adapted. In that order of time, if the model of synchronous generator is used, the inverse time overcurrent protection would not be coordinated precisely.

One possibility for asynchronous wind turbine modelling is shown in Fig. 8Fig. 7: a synchronous generator model connected to the grid through a switch, which opens 2-3 cycles after the fault. Table 4Table 2 shows the results obtained by the different models for tests A and B of the previous example. The behaviour of the proposed model shows a good agreement with the results obtained by PSAD/EMTDC. If the wind turbine contribution is neglected, the instantaneous overcurrent protection operation is not correct. If a synchronous model is used, the inverse curve overcurrent calculation would be wrong.

 

Fig. 87. Proposed model for the protection analysis of asynchronous wind turbines

 

Table 42. Tripping time with the different models

 

Trip time

A

Inverse time

0,400

B

Instantaneous

0,090

 

8.7.3     Coordination problems

The most used protective schemes in distribution networks are fuse blowing and fuse saving schemes. Therefore, an example network is analyzed using both protective schemes.

 

Case 1: Distribution network protected by fuse blowing schemes (Fuse-Fuse Coordination).

 

A fuse is characterised by Minimum Melting (MM) and Total Clearing (TC) characteristics. The Minimum Melting characteristic gives the time in which fuse is melted for a given value of fault current and Total Clearing characteristic gives the fault clearing time of fuse considering fault arc extinction for a given value of fault current.

Fig. 9Fig. 8 shows the example network protected by fuses.

Fig. 98. Fuse blowing scheme

Initially, it is considered that DG is not connected. In this scenario, the network protection is coordinated for a fault on feeder 1 (F1) if both, fuse B1 acts before fuses A1, B2 and C1, and the TC characteristic of fuse B1 is lower than the MM characteristic of fuses A1, B2 and C1 for all range of coordination (Ifmin, Ifmax).

Fig. 10Fig. 9 shows the coordination graphic for fuses A1 and B1.

 

Fig. 109. Coordination graphic of fuses A1 and B1

 

As it has been commented, when DG is connected in a distributed network, the range of fault current (Ifmin, Ifmax) is modified and current flows in two directions (from source side to load side and vice versa). In this new situation it could happen that the system, initially coordinated, losses the coordination.

To illustrate how DG affects system fault current and system protection behaviour, the distributed network showed in Fig 9 is analyzed for faults in different locations.

  • Fault in section BD (Fault F1)

 

For faults in section BD, the current fault levels, Ifmin and Ifmax, increase because of the presence of generators in source side (IB1=IA1+IB2).

In this situation, the coordination between fuses could be lost if the increment of current fault makes the range of coordination (Ifmin’,Ifmax’) exceed the extent of the fuses curves, as it is shown in fig. 4.

 

  • Fault in section CE (Fault F2)

 

Fuses C1 and B2 detect downstream and upstream faults. Moreover, the current seen by both of them is always the same. (IB2=IC1).

In a coordinated system, it is desirable that for faults on feeder 2 (F2), the fuse C1 acts before the fuse B2, and for faults on main line (F1), the fuse B2 acts before the fuse C1. As the current seen by both of them is the same, this requirement can not be fulfilled.

Case 2: Distribution line protected by fuse saving schemes (Recloser-Fuse Coordination)

Fig. 1110. Fuse saving scheme

 

Fig. 11Fig. 10 shows a distribution line where the main feeder is protected by a recloser and the load feeder is protected by a fuse. In this configuration the recloser has to act against temporary faults and the fuse against permanent ones.

For the analysis of this configuration, a recloser actuation sequence Fast-Fast-Slow-Slow is supposed. According to this sequence, if faults occur in the load feeder, the recloser acts opening the feeder breaker according to its fast overcurrent curve. The feeder breaker stays in this state for a defined time until the recloser orders to close it, allowing temporary faults to be cleared. If the fault persists, the recloser acts again. If after the second fast actuation of the recloser the fault is not cleared, the fault is assumed to be permanent. Therefore, after the second fast actuation of the recloser, it changes its overcurrent curve to the slow one, so that the fuse acts faster than the recloser. For the correct performance of the described scheme, the recloser and the fuse must be coordinated, as it is shown in Fig. 12Fig. 11.

Fig. 1211. Characteristics of the fuse (TC and MM) and the recloser (Fast and Slow) for a coordinated system

 

As it is showed in Fig. 12Fig. 11, for a fault in the load feeder, if fault current is inside the limits (Ifmin and Ifmax) the system will be coordinated. From the curves is clear that, in that range of current, fast recloser curve is faster than the fuse one and that the slow recloser is slower than the fuse. If the fuse fails in its actuation, the slow recloser should act as backup protection.

When DG is not connected, the currents seen by the recloser and the fuse for faults in the load feeder are the same so, while current fault is inside coordination range (Ifmin and Ifmax), the system is coordinated, as it is showed in Fig. 12Fig. 11.

Considering that DG is connected, the new situation results into the following changes:

  • The minimum and the maximum fault currents for a fault in load feeder are modified because of the presence of DG.
  • The fault current seen by the recloser is different to the current seen by the fuse.

If fault level increases due to DG, it could happen that fault currents are outside the coordination range. As it is showed in Fig. 12Fig. 11, in this situation the system loses the coordination because the MM characteristic of the fuse is lower than the fast recloser curve.

The different currents seen by the recloser and by the fuse depend on the size, location and type of DG. This difference of current can cause coordination problems to appear, if the current seen by both protections makes the fuse to act before the fast recloser actuation.

 

8.7.4     Solid State Transformer – Sankara

Today’s transformers are single-function devices. They change the voltage of electricity from one level to another, such as stepping it down from the high voltages to lower voltages and vice versa. The new solid state transformers are much more flexible. They use transistors and diodes and other semiconductor-based devices that, unlike the transistors used in computer chips, are engineered to handle high power levels and very fast switching. In response to signals from a utility or a home, they can change the voltage and other characteristics of the power they produce. They can put out either AC or DC power, or take in AC and DC power from wind turbines and solar panels and change the frequency and voltage to what’s needed for the grid. They have processors and communications hardware built in, allowing them to communicate with operators, other smart transformers, and consumers.

 

Smart solid-state transformers are still in the development stage and likely are a few years away from being ready for market—researchers are still working on their efficiency and cost, for example. Taking advantage of their DC capability will require developing new construction standards for homes and businesses.

 

EPRI Moves to Field Demonstration for of the Solid-State Distribution Transformer

Distribution transformers are ubiquitous in the distribution system. Based on a technology developed over 100 years ago, they are extremely reliable in performing their primary function— reducing the electricity supply voltage to a level that can be used safely by customers. EPRI has developed a A prototype transformer has been developed for an “intelligent universal transformer” (IUT) that applies solidstate technology for voltage conversion and provides additional functionality expected to offer distinct advantages in a more complex delivery system, benefitting consumers and utilities. Solid-state technology can improve consumer power quality, provide continuous voltage regulation and reactive power compensation, and facilitate distribution automation. Combined with communications technology, the solid-state transformer becomes a smart node within the smart grid that can help detect metering problems, track asset loading, and serve as a data source for real-time condition monitoring and load modeling. It will also help integrate distributed resources such as energy storage, photovoltaics, and plug-in electric vehicles.

 

Clear Advantages

The IUT converts alternating current (ac) power at various distribution- level voltages to direct current (dc) and ac power ready for residential and commercial use. Unlike conventional copperand- iron transformers, its solid-state high-frequency switching and fast-computing digital control technologies allow it to control and shape its output characteristics. It can provide continuous, accurate control of the voltage levels at every customer location. Because it can regulate the customer-side voltage independently of the distribution voltage through active filtering and line voltage regulation, the IUT will improve ride-through capability for voltage sags and mitigate other power quality phenomena. When combined with energy storage that is connected to the dc tab, it can act as an uninterruptible power supply. Customer voltage control is also becoming increasingly important for utilities as an energy conservation and demand management technique. For future smart grid applications, the IUT can be used to connect distributed renewable generation to the distribution system without the significant distribution voltage variations allowed by conventional transformers. Combining the power electronics required for electric vehicle (EV) charging with the distribution transformer is another promising application. Unlike conventional units, the IUT retains its efficiency regardless of load—a characteristic that becomes more important with the addition of local generation, which reduces the overall load on the transformer. Other beneficial characteristics include:

• The IUT has no liquid dielectrics, eliminating spill risks.

• The solid-state transformer has the capability to convert a single-phase input to a three-phase output, which can be important in some rural areas.

• Solid-state transformers can be built from modules combined to achieve various transformer ratings for kilvolt-amperes (kVA) as well as voltage. But some challenges remain, including matching the life expectancy and cost of traditional transformers.

 

Prototype Development

 EPRI is leading the development Development and demonstration of a fully integrated, production-grade 4-kV- and 15-kV-class solid-state transformer for integrating energy storage technologies and EV fast charging is in progress. The development team includes utilities, power electronics experts, and a transformer manufacturer to provide guidance on taking the technologies from concept to production. EPRI has been working on the IUT’s solid-state technology for a number of years and has completed proof of concept and various prototype designs that are ready to be included in field demonstrations and early deployment, leading to commercialization of the technology. EPRI successfully demonstrated a working IUT in December 2010 at its laboratory in Knoxville, Tennessee, and a field prototype 2.4-kV, 25-kVA model with enclosure, packaging, and high- and low-voltage bushings was deployed earlier this year for evaluation. A fully functional EV fast-charger system was evaluated, along with a variety of communication and performance features. A number of field demonstration projects at multiple host sites are scheduled through 2012 to finalize IUT design, specifications, and manufacturing requirements.

 

 

8.8     Super Conducting networks / A3.16 –Sanjay & Teruo (in process)

 

Imai to add section on TEPCO Super Conducting line demonstration[MGA51]

 

High temperature Super conducting cables is one technology that will enable Transmission of Bulk power over long distances with low losses. “Superconductor cables offer unique power density, efficiency and security advantages compared to conventional power cables and will play a key role in providing the necessary backbone to support the Smart Grid across locations around the world.  HTS wire can conduct up to 10 times the amount of power of conventional cables, which are made with copper wire. They can be placed strategically in the power grid to draw flow from overtaxed conventional cables or overhead lines to mitigate grid congestion experienced in urban centers. They also automatically suppress dangerous power surges to create resilient, ‘self-healing’ Smart Grids that can survive attacks and natural disasters, making them an ideal modernization tool for metropolitan power grids[s52].

 

 

 

8.9     Fault Current Limiters – Paul Mydra

8.9.1     Background

With the growth of the electricity demand, utilities have been upgrading their systems continuously for higher power transfer capability and, consequently, for higher fault current handling capability. Also, the generation system is continuously expanded with more installed capacity, could it be wind or solar farms, nuclear plants, or through other distributed means. A more closely coupled system not only exhibits reduced source impedance values from parallel paths but also an increased number of sources possibly contributing to a fault. Under increased power flow conditions on existing assets, managing fault currents is crucial in order to avoid malfunctioning and damage of equipment as well as to increase system reliability. Increased fault current levels become a source of concern for existing fault current interrupting devices, such as circuit breakers, as the devices become gradually overdutied, and must interrupt fault currents that may exceed the rating it was originally built for. This may create a significant reliability and safety implications on the electric system. 

8.9.2     Reasons for Increased Fault Currents 

There are several reasons for ever increasing fault current levels in transmission and distribution systems. 

Increased Power Transfer Practice – An increasing imbalance between load growth and new transmission line construction combined with difficulties in acquiring rights-of-way, especially for high-voltage transmission network, have imposed unprecedented demand on the power delivery system. In order to meet this increasing electricity demand, utilities have been pushing more power through the existing system resulting in higher fault currents. 

Meshed Networks- Present networks are getting more interconnected for the purpose of enhancement of reliability and flexibility in the power transmission. A more closely coupled system not only exhibits reduced source impedance values from parallel paths but also an increased number of sources possibly contributing to a fault.

New Generation – The addition of new conventional generation (independent power producers and co-generators) and distributed generation (DG), such as thermal solar power and photovoltaic systems, wind generators, fuel cells, micro-turbines, and combustion turbines etc., to existing generation is constantly increasing. Such addition in generation in an existing system increases the fault current level throughout the existing electric system. In some cases, especially when the system has high penetration level of such generations, the fault current could be higher than the existing interrupting capability of interrupting devices (e.g., circuit breakers), which may create a safety hazard.

 

8.9.3     Impact of FCL on protection - Johann

The management of power systems in countries in all parts of the world is changing nowadays and there is a strong tendency towards separating generation from transmission. In this deregulated environment the utilities responsible for operating the networks are losing control over the sitting and scheduling of generation. Moreover, the connection of independent power producers to transmission, sub-transmission, and distribution networks causes an increase of short-circuit currents not included in previous long-term planning forecasts.

A consequence of this development is that in certain parts of the networks the short-circuit currents approach, or even exceed, the allowable values based on equipment rating, primarily circuit breaker interrupting capability. The problem of excessive short-circuit currents has therefore become an import issue for the operators of power systems and there are clear indications for a growing interest in fault current limiting (FCL) devices rated for applications in the high voltage system. But up to now, with a few exceptions, there has been relatively little progress in developing suitable devices and bringing them to the market. Only in recent years have several successful field and large scale laboratory tests with novel FCL devices been conducted which indicate significant advancements towards commercial applications. However, before FCLs will be deployed in the utility market the interaction between these new devices and the power system network, in particular the existing protection system, must be well understood.

The document ([22]) performs a guideline to cope with this problem. Starting with a limited scope on typical medium voltage system structures this guide first reviews the most relevant system protection techniques such as overcurrent, distance, differential, and directional protection in the context of FCLs.

 

This report establishes qualifying criteria for the different FCL technologies with respect to the application of system protection. Of particular relevance is the FCL behavior in the network with respect to the relay measurement function. The first cycle and follow current behavior has to be discriminated as shown in Figure X.

Figure X: Typical fault current wave shape and characteristic data:

 

Figure Y shows the simulated time domain waveform of a thyristor-based solid state FCL (SSFCL) which utilizes phase angle control to maintain a through fault current for protection coordination. It becomes evident that the signals of three differently averaged current values (True RMS, Peak/sqrt(2), and Fundamental) for the distorted current waveform are not only different at the steady state follow current but also exhibit different time evolutions within the first cycle.

 

Figure Y Simulated time domain current signals due to phase angle control of a solid state FCL (SSFCL): line current and three differently averaged current values of the distorted current waveform

The conventional protection zone concept is reviewed and applied to differentiate the various locations of faults with respect to the FCL location. This is the first step to establish a framework which correlates the influences of different FCL measures, characterized by their previously discussed qualifying criteria, to the various protection functions. A series of examples are discussed in detail to illustrate the new framework.

It is concluded that while FCLs will, in general, interact with the existing protection system the presented methodology helps to clearly identify the specific conditions for which such interactions occur. This framework can be generally applied since it is not restricted to the example cases presented in this document.

([23]), ([24]), ([25]), ([26]), ([27])

8.10 Evolution of Communication Technology

8.10.1 Communication requirements for coordinated protection and automation functions

(supports, acceptable delays, reliability architectures) – Authors for above item should estimate the communication performance for each function

  •  

8.10.2 Wide Area Communications - Adam

The monitoring and control of the future grid will require fast, secure, and wide-area messaging.  IEC 61850 defines a fast communication mechanism known as the Generic Object Oriented Substation Event or the GOOSE.  As the name implies, the original application for the GOOSE was envisioned to be inside the substation.  A GOOSE message has no IP address so when this message enters a router, the message is normally discarded (note: there are exceptions as some routers can be manually configured to map a GOOSE message into a routable packet, however, there are inter-operability issues).
 

As functions such as Synchrophasor Data Streaming, Wide Area Control, Demand Side Management, Distributed Generation Regulation, etc. become main-stream, there arises a requirement to be able to perform wide-area publish-subscribe communications.  A new IEC 61850 profile, known as IEC 61850-90-5, has been developed to meet identified as well as future functional requirements.


IEC 61850-90-5 profile starts with the concept of a Dataset – as defined for GOOSE and Sample Values (SV) – and wraps the respective dataset in a UDP/Multicast IP wrapper.  The “routable” versions of GOOSE and SV are known as R-GOOSE and R-SV respectively.  Similar to GOOSE, the R-GOOSE is sent on detection of change of state – as well as periodically in a keep-alive mode of operation.  The R-SV definition of the profile was designed to address the function of Synchrophasor Streaming (similar to the streaming of sample data) over a wide area.

 

The Multicast IP address enables the message to be automatically routed from one to many wide-area locations.  In addition, the profile defines security mechanisms for Authentication (Secure Hash Algorithm – SHA 256) and Encryption (via the Advanced Encryption Standard – AES).  To provide dynamic updating of the Authentication and Encryption security key, a point-to-multipoint Key Exchange mechanism, based on the RFC 3547 Group Domain of Interpretation, is defined.

 

In Eskom, the South African Power Utility, the following media are currently utilized for teleprotection requirements:

  • Eskom Bandwidth Management Equipment (BME) network;
  • Fibre optic cables;
  • Power line carriers.

Most of Eskom’s telecommunication requirements and infrastructure is dependent on the Eskom BME network which consists of a number of microwave radios and fibre optic cables.  Most of the fibre optic cables belong to Eskom.  Due to increasing demand for bandwidth and infrastructure ageing, Eskom is currently replacing the BME equipment.  The transport network will be migrated entirely to OPGW fibre optic and in the access network a mix of fibre optic cables and radio links will be used.  Where no OPGW fibre alternative exists, Wrap, ADSS and trenched fibre optic cables will be deployed.

The refurbishment will provide a substantial increase in network bandwidth to improve SCADA capabilities, prepare for Smart Grids unified communications and connectivity for IT requirements.  In the process a migration from ring to a meshed network topology will be achieved.  Where acceptable, IT and Operational Technology (OT) networks will be consolidated, with due consideration to OT stringent dependability, security, latency and availability requirements, risk profile and legislative requirements.

It is the intention of Eskom to migrate its telecommunication network to a fully packet - based OT IP/MPLS network in the future following rigorous testing, evaluation and acceptance processes.  Currently for differential protection only dark fibre optic is used, for impedance protection either PLC, BME or dark fibre optics can be used.

The new telecommunication system should meet the following performance requirements for Transmission teleprotection purposes:

  • Availability = 99.9 %
  • Latency = 5 ms or less
  • BER = 1x10-6

 

8.10.3 Cloud Computing[VL54] - Valeriy

23/10/2012[VL55]

 

Cloud computing is novel paradigm of growing importance and popularity. It refers to the use of computing resources that are delivered as a service over a network. Cloud computing relies on sharing of resources (hardware, software and data), through many available techniques such as Infrastructure as a service (IaaS), Platform as a service (PaaS), Software as s service (SaaS), Test environment as a service (TEaaS).

Cloud computing can facilitate many SmartGrid functions which require high performance computations and modelling, or ample data storage for the data generated by measurement instruments such as smart meters. It can also provide an appropriate infrastructure for real time control or coordination of smart grid ([28]).

There can be also an adverse impact of cloud computing to power systems: The future proliferation of cloud computing will imply constructuion of larger computing facilities and data centres closer to the consumers, i.e. at the level of Internet Service Providers, or at the (or close to) large enterprises. These installations will consume huge power with the load non-uniformly distributed during the day.

 

 

9.     Developments in Protection and AutomationTechnology

9.1     Evolution of SAS Architectures

 

  • Substation SAS Architecture  - Robin – Mark and Sankar supporting – 
  • Distribution of functions
  • Augmented functionality of future IEDs
  • Process Bus
  • Distribution Automation Architecture


- Transmission – Mark to provide drawing and words


-Distribution – Mika

 

Future redundancy[VL56] : Redundancy through processs bus – including time sync redundancy - Mark

 

9.2     Migration of synchrophasors to the distribution grid - Mark


 

9.3     Coordination between Primary and Backup protection – Andrei / Jorge Cardenas / John Cuifo

 

9.3.1               Distribution / Sub-Transmission (<69kV) [Ciufo][VL57]

 

The fundamental protection and automation strategies used in operating power and distribution system has historically changed very little over the decades.  Protective devices where  designed to operate at pre-established fixed thresholds based on locally sensed quantities in response to anticipated contingencies.    However more recenting, these strategies are changing to support current and much needed new power system initiatives.

 

As described above in Section [6.1.1]  the utilization of renewable sources of energy are leading to a growing need for enhancing operating flexibility in power systems.  Technical advances in generation technologies permit production of generators that are modestly rated in comparison to large-scale utility generation.  They tend to be more economically placed closer to loads, and hence best integrated and distributed DGs at sub-transmission/distribution voltage levels. Moreover, there is a demand to connect an increasing number of  DGs on a given feeder.  Regulatory requirements to improve reliability is causing the  industry to install more in-line reclosers for sectionalizing.  The need to introduce cost-of-use billing has resulted in the replacment of legacy revenue meters with smart meters.  Asset managment inititives has promoted the installion of asset equipment monitoring to support asset replacement programs.  These current inititiatives, and others, are demanding more from sub-transmission and distribution systems and forcing them to  become more ‘smarter,’ and therefore, protection and automation strategies will need to change and become more advanced then they where historically.

 

These developments will require protection and control devices and schemes to rely on a more global view of system operating conditions, which will require greater communication between devices and/or a central hub.  This approach leverages the communications infrastructure capabilities contemplated  for programs such as distribution automation and smart (self-healing) grid as discussed in Section 7.9.3.

Traditional feeder protections in North America use a series of non-directional overcurrent protective devices to provide and co-ordinate fault coverage.  As discussed in Section 8.6.3, legacy feeder protections, based on economics, used simple overcurrent relays, fuses, and reclosers.  These protection devices coordinate to isolate the faulted section by means of tripping respective fuses, reclosers or the feeder circuit breaker.  Protection systems are designed to be selective which means that they first trip the isolation device that result in the minimum system disruption.  If that doesn’t isolate the fault (generally this would happen if their is a failure or a permanent type of fault) then it must select  the feeder breaker to trip, to isolate the fault and typically this will result in lose of all connected feeder customers.

 

The philosophy of protection design is to divide the distribution system into protective zones that can be protected adequately with the minimum amount of system disconnected. These protection zones should cover the entire distribution system completely, leaving no part unprotected.  When a fault occurs the protection is required to select and trip only the circuit/section  needed for fault isolation.   This feature of a protection system is referred to as coordination or selectivity. Selectivity is the ability of a relay to differentiate between conditions for which it must act or not, to initiate an action.  In doing so, for a given fault, the minimum number of interrupting devices operates to isolate the fault such that service is interrupted to only a minimum number of customers.   Relays must be able to recognize abnormal conditions on their own protected element and generally ignore abnormal conditions outside their protected area or zone. Discrimination methods such as: measured quantities, time, combinations of magnitude and time, distance, etc., are used to provide selectivity/coordination.

 

Distributed Generation can disrupt coordination of these devices by altering the normal fault current distribution in a manner that depends on the type of the DG resource as well as its relative location – refer to Section 8.6.3.  As a consequence, protections may either over-reach or under-reach (i.e. become overly sensitive, resulting in unnecessary customer outages, or be desensitized such that fault clearing is unduly delayed, thereby risking equipment and safety).

 

Future distribution protection systems and coordination requirments will need to migrate from traditional simple non-communication based, to more sophistaicated and transmission like schemes.  The new distribution protections will need to satisfy current automation strategies and meet more advanced requirments, to do so the following are envisioned:

 

  • Underpinning these new distribution objectives is the need for reliable communications.  This will allow more zone-tight coordination needs and also to provide more advanced smart grid functionality
  • A strategic approach to smart grid technologies is to develop smart grid functional interface modules for each of the basic distribution building blocks.  Each module can be repeated as required, examples are developing a smart transformer load station, a smart distribution station, a smarter recloser, etc. For each module a set of functions  are made available  so that the module can operate as a standalone entity and also  meet any modular functionality  requirements to meet smart grid system needs such as self-healing, monitoring, etc.
  • Simple overcurrent relays will need to be replaced with distance type relays.  This will improve sensitivity, directionality, and loadability.  Moreover, electromechanical relays should be replaced with Intelligent Electronic Devices that provide an integrated platform for multiply protection and automation functions. The same should apply to reclosers.

 

One possible c0ordination scheme is depicted below assuming point-to-point communication availability, and using a directional blocking schemes. With improvments in distribution communication such as band-width, and support of such protocols as IEC61850, will allow further enhancements. Further distribution coordination will demand more sophistication and reliance on communication to meet higher distribution power system demands and performance.

 

 [MGA58]

 

 

 

 

9.4     Use of IEC 61850 and future extensions  - Emiliano[MGA59]

Cloud Protection – Alex

Other abstract Logical Nodes (distilled form section 8 above)

Grid Connection Policy needs

Need for device models

 

Organization structure

Organizational Functional Model and associated skills

 

9.4.1               Future system engineering tools  - Valeriy[VL60]

IEC 61850 paves the way to systems engineering supported by software tools of new generation, covering the entire design process from specification to geberation of code and its deployment. One such tool is BlokGrid, developed at the University of Auckland by the group of Prof. Valeriy Vyatkin.

As shown on Figure Y1 BlokGRID consists of 3 main modules: system configurator, generation module and co-simulation.

BlokGRID presents both system and IED configurators in one tool, adopting the IEC 61850 system engineering process. The first module represents IEC 61850 system configurator. It includes three components: SLD, SSD/SCD and schema editors.

Using a database of symbols, the engineer designs the SLD in the SLD editor. In the SSD/SCD editor, customer requirements are represented using LN concepts and a database of IEC 61850 LNs. Moreover, BlokGRID provides the ability to easily sketch automation, control and protection schemes on top of the SLD as if using paper. This functionality is implemented in the schema editor. Then the engineer can proceed with detailed system engineering.

Figure Y1. BlokGRID architecture

BlokGRID can parse the system design to identify physical components, functions and function interactions. This is done in the "Generation" module of the software. Using a database of iLNs, the tool generates an executable IEC 61499 application directly from the specification. The code editor allows modification of the generated user-defined automation, control and protection logic. The tool generates the corresponding system configuration with the defined network topology and device set. Signal engineering while allocating functions to devices (distribution) is supported by automated configuration of the detailed data flows between IEDs, and allocation of communication addresses (GOOSE datasets, signal inputs to clients, sub-networks). A custom library of iLN FBs is created according to the DataTypeTemplate section of the SCL file. BlokGRID generates a Matlab simulation model of the system from the SLD contained in the SSD and SCD files.

BlokGRID can import ICD and CID files, It can also export CID files containing the necessary information from the SCD file of the designed system in order to configure the IEDs. BlokGRID can also generate target files containing code – programs for IEDs and/or automation controllers.

In the co-simulation module, BlokGRID performs functional simulation of the designed control logic in conjunction with the utility model. The automation, control and protection code can be downloaded to the simulated devices or deployed to real hardware. The tool provides a real-time, event-driven simulation via a GUI, e.g. an engineer can introduce a phase-to-ground fault into the model of distribution network. Additionally BlokGRID supports hardware-in-the-loop testing with a controller, protection device or other equipment. The developed and validated code now can be deployed to the hardware.

Thus, BlokGRID supports the system design process from drawing a SLD and capturing customer requirements through to the deployment of the validated automation, control and protection solution to physical devices. Throughout the process BlokGRID maintains compatibility with other standards-based tools, by supporting standardised file formats according to IEC 61850 (SSD, SCD, ICD, CID and IID).

 

9.5     Adaptive Protections

9.5.1     Definition of Adaptive Protection

Adaptive Protection has been defined as follows ([29]) ([32], S.H. Horowitz, A.G. Phadke, J.S. Thorp, 1988, “Adaptive Transmission System Relaying”) : "Adaptive Protection is a protection philosophy which permits and seeks to make adjustments in various protection functions automatically in order to make them more attuned to prevailing power system conditions". This concept has been formulated several decades ago, but its implementation in networks and substations was somewhat slow, due to technological obstacles at the beginning and certainly also due to the fact the control- and protection engineers often do not feel comfortable with automatically changing protection settings and -parameters. In addition to that, the relatively static character of networks and sufficient operation margins did not push to implement adaptive functions in the protection or control systems of substations. The actual penetration of Adaptive Protection and Control in networks is difficult to estimate.

Today, the demand for higher sensitivity and lower operation time is growing. Lower safety margins and higher constraints can be anticipated in Future Networks with the application of Adaptive Protection. The acceptance of this solution is growing. The technology has been proofed in analogue and, even more, in microprocessor devices. The adaptive algorithms have been implemented in protection by manufacturers. Now, the technology comes to the level of a protection engineer: Adaptive Protection using programmable logic in IEDs and substation-wide or system-wide adaptive functions using IEC 61850 or another communication standard.

There are basically two types of adaptive functions.

Continuous Adaptive Functions: Depending on the actual observation, continuous adaptive functions automatically change their settings in a continuous way in order to block unnecessary operation or extend the protected zone.

Discrete Adaptive Functions: Discrete adaptive functions switches between pre-selected settings and modes of operation automatically in response to power system changes.

 

 

9.5.2     Possible applications of Adaptive Protection in Future Networks

([29]) lists a number of applications for Adaptive Protection and Control. Some of these applications have since been described more in detail. Some other applications have been added. Most protection schemes require from extensive information on network and system state (binary signals), other functions take advantage of using available information intensively: spectral analysis, account for pre-fault conditions.

Globally, the applications of Adaptive Protection and Control that may answer to needs identified for the Future networks include the following items:

  • Protection setting adaptation to the network and system state
    The protection system is designed and parametered in a way to guarantee a selective fault elimination in nominal conditions. This frequently leads to a higher fault elimination time when selectivity is achieved by tripping time of protections. This is often the case for overcurrent protection in radial systems or for distance protection in meshed systems.              
    In cases where the network is operated under stressed conditions, high tripping times may lead to out-of-step operation of generators or even to a risk for system blackouts. In these cases, it may be preferable to achieve faster tripping even at the expense of loss of selectivity. The choice of protection settings depending on the network conditions (topology, breaker state) is the principal application of Adaptive Protection. Another approach is the continuous automatic adaptation of settings to the measured voltages and currents in the network, when certain factors are compensated in the protection in order to achieve maximum sensitivity and speed of protection. This effect is best observed in distance protection with adaptation based on super-imposed components.
  • Conditional network restoration
    During network restoration after a partial or complete blackout, particular protection settings or functions may be needed. These features can be implemented on the base of Adaptive Protection or Control. Modification of the harmonic restraint for certain transformer protections is one example of an Adaptive Protection approach of this type.
  • Undervoltage / Overvoltage control
    In deteriorated network conditions, overvoltage or undervoltage situations may occur. In order to achieve a stable situation, it may be desirable to change over- or undervoltage settings of protections or automations under certain conditions. Undervoltage tap changer blocking of transformers is one example of an Adaptive Automation approach of this type.
  • Distributed Generation
    The amount of Distributed Power injected in a line or a network may require Adaptive Protection or Control algorithms. This may concern the ability of a subnetwork to operate as isolated network, adaptation to the power flow in a feeder, adaptation of the neutral point grounding of the network to keep Zd/Zo in a predetermined range, or other similar applications. This type of algorithm may take the available information about network topology into account.
  • Adaptive Overload Protection
  • Parallel Lines
    Performance of distance protection on parallel lines is negatively affected by mutual coupling effect between lines. The use of multiple locally available signals applying adaptive techniques enables to improve the protection performance ([30]).
  • Adaptive single-pole and multi-pole reclosing
    The reclosing philosophy after tripping may have similar constraints than the protection: depending on the network state, it may be better to systematically reclose after fault (nominal conditions) or to reclose only under certain conditions in order to limit impact on a reclosure on fault.

 

([31]) distinguishes two types of Wide Area Adaptive Protection and -Control algorithms :

  • Anticipation of network vulnerabilities and adaptation aiming at making the network more robust for the identified issues.
  • Adaptive System response to failures and incidents.

Most applications identified above can be used in either of these two philosophies.

([32], S.H. Horowitz, A.G. Phadke, J.S. Thorp, 1988, “Adaptive Transmission System Relaying”) 

 

9.5.3     Issues of Adaptive Protection Application in Future Networks

The adaptive protection application will face the following challenges in Future Networks: setting calculation, algorithms performance. New generation techniques (wider usage of renewable energy sources and distributed generation) along with the flexible control of power transmission will change the nature of transients. Switching will occur more often as a means of power flow control. Current reversal conditions will come down to distribution networks. The time constant of the electro-mechanical transients will decrease. Due to these new constraints, the adaptive protection may need changes in algorithms based on super-imposed components. Moreover, the settings calculation and verification may need automation, as the number of possible network states and conditions increases dramatically. This can be done with the aid of real-time and off-line simulation tools and adaptive protection models. The protection scheme should be tested against all possible conditions.

Another problem that adaptive protection should address today is non-linear behavior of regulators and semi-conductor converters (e.g. inverters or wind turbine generators). If adaptation to the load current is based upon super-imposed components of the fault conditions to the pre-fault network, the algorithm may have to be enhanced or replaces by new approaches. As the share of non-linear primary elements in the power system grows, there is a certain limit of the application of super-imposed components. Accordingly, adaptive protection may mal-operate or display low sensitivity during a fault due to inconsistency of the fault components.

 

 

9.6     Adaptive tripping on phase-to-phase faults – Alex

[MGA61]

 

 

9.7     Advanced Adaptive reclosing

 

Adaptive reclosing is a crucial part of the future network. If a fault occurs on a transmission line, the line is tripped at all feeding ends in order to isolate the fault. According to the accumulated statistics, 70-80% of all faults in overhead lines disappear after the disconnection from a power source. Therefore, autoreclosing is a widely accepted means of normal power flow restoration. Besides, the systems on the line ends may go out of step due to a major disturbance and the following disconnection of the line. One of the systems accelerates, while the other one decelerates. The longer the transmission is disconnected, the higher the transmission angle and the frequency difference. As a result, the transmission angle may reach high values at the instance of circuit breaker closing. Thus, the dead time during autoreclosure should be as low as possible to prevent stability violation. However, closing onto a fault is far greater disturbance for a power system. Thus, the delay is normally calculated as the absolutely maximum time for fault clearance in the most unfavorable weather conditions.

One of the possibilities to decrease the dead time of autoreclosure is an adaptive delay before closing of the circuit breaker. For this purpose, the line is monitored for a persistent fault. The line is enegized only if the fault is cleared. The actual technique for fault detection depends upon line structure and operation modes.

 

9.7.1     Adaptive single-pole autoreclosure

UHV networks 300-750 kV are used for power transmission for a long distance. The corresponding lines are mainly equipped with shunt reactors and operate with single-pole tripping and autoreclosure. A phase-to-ground fault is tripped on both line ends, but it is likely that an arc is fed by the healthy phases electromagnetically. The shunt reactors form a complete circuit together with the fault. A quick and reliable autoreclosure is secured by conditional autoreclosing, when circuit breakers are closed conditionally. There are two functions employed:

  • fault clearance detection,
  • successful energizing detection.

The fault clearance detection is made by a special impedance relay and voltage relays. Then, the line is closed and thus energized. This is detected on the other line end with a voltage (sometimes polarized) relay, and the second breaker is closed. The total dead time can be less than 300 ms rather than 500 ms and more in case of fixed dead times.

 

9.7.2     High-frequency probation

A line with three poles open can be verified against a permanent fault using high-frequency probation systems. A high-frequency signal may be of low amplitude and energy. A signal reflects from the fault (as it offers a different natural impedance) and returns to the point of observation. The time of the signal propagation forth and back is measured and compared to the setting. There are more sophisticated high-frequency techniques that enable estimation of a distance to a fault or fault parameters.

 

Adaptive autoreclosing impact upon Protection and Automation Systems Adaprive autoreclosing may either accelerate or delay closing of the transmission line. Therefore, protection and automation systems must be compliant to any scenario offered by adaptive autoreclosing. This implies that protection must take into account certain time range of the pause.

The minimum boundary should be coordinated with the tripping impulse length and the protection reset time after a major fault. Circuit breaker failure protection should have sufficient time to reset in order to prevent unwanted operation.

The maximum boundary should be coordinated with the autoreclosing cycle counter reset. Another coordination is made in the out-of-step protection in order to prevent asynchronous switching of the transmission. If the transmission angle reaches critical values (stability issue), closing should be cancelled.

9.8     Sensing technology – T.W. Cease

 

Sensing technology will need to be divided into transmission substation technology and distribution technology. For distribution substation current and voltage sensing the low cost requirements will keep many of the alternative (unconventional) technologies from being implemented for some time. With wider spread implementation in transmission substations the cost will come down with time. For transmission substations there are a number of sensing technologies available. The conventional technologies all have limitations but they are well known and work around solutions are available. Due to cost factors of current technologies voltage sensing at UHV is performed by CVT. There are accuracy and reliability issues that can arise. In order to solve many of the technical and cost issues there are becoming available unconventional sensors. There are available optical based current sensors and rogowski coil current sensors ([34]) ([35]). Both have technical and cost advantages over conventional sensors but their output does not match the requirements of traditional relays. With the advent of the process bus and merging units that problem has been solved. However that is a large amount of convention equipment installed and a wholesale change out would be cost prohibitive. Optical voltage sensing exists and has numerous advantages over conventional voltage sensor at UHV and EHV. Again the main problem with implementation is that the output from the optical sensor is different than that of a conventional sensor. This problem can again be solved by use of the process bus and merging units ([36]) ([37]).

 

9.9     Application of SIPSUse of synchrophasor measurements - Mark[MGA63]
- Include application of Synchrophasors for Distribution State Measurement

- Synchrophasor-based Linear State Estimation

 

 

9.10 Faster than real-time contingency analysis – Paul Myrda - Marek[MGA64]

 

 

 

9.11 Earthing Policy changes – Jorge Cardenas[VL65]

  •  

9.11.1 Introduction

 

Tests have shown that ungrounded systems are subject to transient over voltages which can be caused by normal switching, switching of a circuit containing a ground, or by repeated restrike of an arcing line-to-ground fault. Highest voltages are obtained when arcs are involved in the circuit disturbance. These over voltages may have magnitudes up to six times the normal line-to-ground voltage. The line-to-ground insulation on all equipment connected to the system is also subjected to these over voltages, which may weaken it so that it will eventually fail. It is not uncommon to experience multiple instantaneous equipment failures on different parts of an ungrounded system when this condition occurs. Due to standard insulation practices, this condition is more likely to occur on system voltages of 2,400 volts and above.

 

One important advantage (in some applications) with an ungrounded system, is that the phase-to-ground fault current is negligible and does not result in load interruption. However, if nothing is done to clear the fault, a safety hazard exists. In the meantime, if a fault occurs on another phase on any feeder in the system, relatively high line-to-line currents could flow, which would result in equipment damage and tripping of one or both feeders (because there will be a phase-to-phase fault under these conditions). The recommended operating procedure for an ungrounded system is to install some type of ground fault indicator using lights and/or audible alarms when a phase-to-ground fault exists.

 

However, high resistance grounding can be supplied so that the ground current is not appreciably higher than the capacitive charging current, and zero sequence relaying can be applied to indicate fault location. This is one of several schemes available to systematically and promptly locate the ground fault without interfering with production. This concept is desirable when system shutdowns are intolerable.

 

Grounded systems are designed to limit the transient over-voltages to values within the capabilities of equipment insulation and thereby extend equipment life.

 

Some benefits derived from system grounding are:

 

1. Greater personnel safety.

2. Improved service reliability.

3. Reduced transient over voltages and equipment fault protection.

4. Improved lightning protection (when solidly grounded).

5. Reduced operation and maintenance expense.

6. Automatic location and isolation of circuits that develop ground faults.

 

9.11.2 Ungrounded Systems

Ungrounded system is an electrical power system that is operating with no intentional

ground connection to the system conductors are generally described as ungrounded. In

reality, these systems are grounded through the system capacitance to ground. In most

systems, this is extremely high impedance, and resulting system relationships to

ground-to-ground are weak and easily distorted.

 

Two principal advantages are attributed to ungrounded systems. The first is operational: the first ground fault on a system causes only a small ground current to flow, so the system may be operated with ground fault present, improving system continuity. The second is economic: No expenditures are required for grounding equipment or grounded system conductors.

 

9.11.3 Solid Grounding

Numerous advantage are attributed to grounded systems, including greater safety, freedom from excessive system over-voltages that can occur on ungrounded systems

during arcing, resonant or near-resonant ground fault, and easier detection and location of ground faults when they do occur.

 

A system is considered to be solidly grounded when its neutral is connected directly to a station ground or earth with no intentional impedance in that connection. Ground fault currents on solidly grounded systems are about equal to three-phase fault currents, and fuses or relays will operate satisfactorily, where very high fault currents can occur, burning of the core iron may occur under short-circuit conditions.

 

Since motor circuits are very susceptible to ground faults, the impact of motor operation on system voltage is an important criteria for determining the method of grounding. Another important factor to consider is the method of ground relaying, if reduced ground fault current is desired.

 

Low-voltage systems (600 Volts and below) are usually solidly grounded (when grounded) because most low-voltage protective devices (circuit breakers and fuses) are phase-type devices, and require high currents to operate. New static devices do permit ground detection at much lower current values than phase trip devices. Low-voltage systems that use power circuit breakers can be relayed in a manner

similar to high-voltage systems. Since low-voltage circuits are much more numerous, the increased cost of the additional equipment could be prohibitive, which is why phase over-current clearing (fuse or circuit breaker) is usually employed when the system is solidly grounded.

 

Some restrictions of Solidly grounding:

 

a) Grounding of solidly grounded neutral systems shall be accomplished in a manner

illustrated by Diagram 2740 [PTM66]if the the system neutral is available at the service entrance

equipment.

 

(b) If the grounded neutral conductor is carried into the premises, it shall be identified

as the grounded conductor. The minimum insulation level for neutral conductors of

solidly grounded systems shall be 600 volts.

 

EXCEPTION: Bare conductors be used for the neutral of direct buried portions of

solidly grounded systems.

 

(c) The neutral grounding conductor shall be permitted to be a bare conductor if

properly isolated from phase conductors and protected from physical damage.

 

9.11.4 Resistance grounding[MGA67]

 

9.11.4.1 High resistance grounding

Condition of High resistance grounding:

 

1. The conditions of maintenance and supervision assure that only qualify person will service the installation.

2. Continuity of power is required.

3. Ground detectors are installed on the system

4. Line-to-neutral loads are not served.

 

The resistor of a high-resistance scheme is sized according to the following

steps:

 

1. Measure the charging current or estimate the capacitive reactance of the system.

2. Select the size of the grounding resistor such that it is equal to or slightly less than

one third of the system ie.

 

  or  

 

This will provide a low fault current to minimize damage, yet it will limit transient over voltages to less than 2.5 times the normal crest value to ground (when it is under a no fault condition, the ground resistor will bleed off the system capacitance thus limiting the over voltages.)

 

High resistance grounded (distribution-transformer grounded):

 

In this method, the main generator neutral is connected to ground through the primary of a single-phase transformer. A resistor is connected across the secondary of the transformer to provide a high-resistance neutral ground connection. The transformer and resistor are sized to produce an equivalent ground resistance numerically equal to or less than the total three-phase capacitive reactance to ground of the generator and other equipment connected to the generator bus. Most unit-connected generator are grounded in this manner. Using this type of scheme, the fault current is typically limited to 5-10A.

 

9.11.4.2 Low resistance grounding (Neutral resistor ground)

This method is used when generators directly connected to the system without a step-up transformer. It permits a higher level of fault current, which is generally several hundred amperes to about 150% of rated machine current. It permits sufficient fault current to operate the differential relays for all machine faults except those near the machine neutral. This resistor should also meet dielectric requirements of full phase-to-ground voltages or better.

 

9.11.4.3 Low inductance Grounding

This grounding method is using a reactor connected between the system neutral and ground. Since the ground fault may flow in a reactance-ground system is a function of the neutral reactance, the magnitude of the ground fault current is often used as a criterion for describing the degree of grounding. In a reactance grounded system, the available ground fault current should be at least 25% and preferably 60% of the three-phase fault current to prevent serious transient over-voltages (X010X1). This is considerably higher than the level of fault current desirable in a resistance grounding system, and therefore reactance grounding is usually not considered an alternative to resistance grounding.

 

In most generators, solid grounding, that is, grounding without external impedance, may permit the maximum three-phase fault current that the generator can deliver and for which its windings are braced. Consequently, neutral-grounded generators should be grounded through a low-value reactor that will limit the ground fault current to a value o greater than generator three-phase fault current.

 

In this case of three phase four-wire systems, the limitation of ground-fault current to 100% of the three-phase fault current is usually practical without interfering with normal four-wire operations. In practice, reactance grounding is generally used only in this case and to ground sub-station transformer with similar characteristic.

 

9.11.4.4 Resonant grounding:

 

This method can be used for the unit-connected generators. The main purpose of this method is to minimize phase-to-ground fault current. The power dissipated in the effective resistance should be equal to or greater than the three-phase zero-sequence capacitive reactance of the generator bus.

 

 

This method of grounding is used primarily on systems above 15kV, consisting largely of overhead transmission or distribution lines. Since systems of such constructions are rarely used in industrial or commercial power systems, the ground fault neutralizer finds little application in these systems.

 

The main benefit of this system is the natural shelf-extinguish of the fault when the voltage cross to zero, avoiding unnecessary operations of the circuit breaker, being an interesting application in rural or in radial distribution networks.

In general, protection is only used for permanent faults and the analysis will refer only to this specific case

Figure 1

 

Figure 2

 

In Figure 1 and 2 we have the two possible arrangements for Petersen Coil Grounded, also called “Resonant Grounded”. In Figure 1, because there is no Neutral point, it is needed to create an artificial Neutral using a ZigZag Transformer. In Figure 2, MV side is connected in Y, for that, we use the Neutral Point on the secondary side of the transformer to connect the Petersen Coil

In order to identify the phenomenon, there were made circuit analysis using the EMTP-ATP Software (transient analysis). In Figure 3 we can observe the circuit used and the parameters of the network. We have a HV side of 150 kV and a MV side of 24 kV. The Transformer is DY11 grounded with a Petersen Coil. We can observe also the resonant point calculation in order to determine the Reactance value.

In Figure 4, we have the same circuit, but with the fault point for Phase-to-ground simulation

 

[MGA68]

Figure 3. Parameters and Calculation of the Reactance Grounding

Figure 4. Circuit and Fault Point for Analysis


Analysis Results

Self- Extinguish Faults

(a)

b)Transient non self-extinguish faults

(b)


c)Permanent Faults

(c)

Figure 5. The most common Fault Conditions. a) Self-extinguish faults, b) Transient non self-extinguish faults, c) Permanent Faults. In the three Figures the order is as follows: Current on the faulty phase, residual current circulating through the Petersen coil and residual voltage in the Petersen Coil

 

9.11.5 Earthing Policy during islanding

 

 

9.12 Implications for Testing - Alex[MGA69]

 

 

 

10.           Future Network Use Cases

 

Characteristics of Electric Energy Market[VL70]

 

Item

Nowadays

In Five Years

In Ten Years

Scheduling and dispatch

Centralized and based on computational models (weekly basis)

Centralized and based on computational models (daily basis)

Centralized and based on price statements by generation utilities and consumers (daily basis)

Ancillary services

Mandatory supply

Mandatory supply

Purchase by operator through price indication and payment by those who caused the unbalance

Short term price formation

On weekly basis via computational models

On daily basis via computational models (24 hours of next day)

On daily basis (48 half hours of next day) via price statements by generation utilities and consumers

Demand side management

None

Resource for peak supply and achievement of "target-level"

Resource for peak supply and achievement of "target-level", taking part in short term price formation

Demand reaction to short  term price

Practically none, except via arbitrament

Intelligent meters will lead to hourly basis metering

Intelligent meters will lead to half-hourly basis metering

Reliability individual purchase

None

Smart grids will lead to selective load shedding

Price of selective load shedding will be part of the contract of energy/power supply (less expensive loads are disconnected first)

Interruptible generation management

No large scale application so far

Intelligent grid adaption based on spinning reserve as regulating resource (AGC integrated to smart grids)

Use of intelligent grid for managing AGC in the coordination of interruptible supply (demand creation in case of excessive generation)

Synchrophasor Measurement (WAMPACS)

No large scale application so far

Use of WAMS and WACS for on-line parameterization of power system control devices (PSS, voltage and speed regulators, FACTS, etc.)

Maximization of transmission capacity by means of WAMPACS

Expansion of energy offer

Mostly by regulated market

By both regulated and free markets

By both regulated and free markets

 

10.1 System instability – BR

 

Power system stability can be defined as the ability of a power system to sustain an active power transfer in normal operating conditions and to reach a feasible steady state after being subjected to a disturbance. Typical disturbances are: short circuit in an important power system component, sudden loss of large blocks of load or generation shutdown, cascading events, etc. The power system becomes unstable when stability limits are violated.

 

NERC [1] defines Total Transfer Capability as the smaller value among thermal limit, voltage limit and stability limit. Thermal and voltage limits may be defined by off-line simulations, but defining stability limits is a more subtle problem. Thermal and, to some extent, voltage limits can be even violated during short periods of time. The stability limit is a function of the power system state vector, meaning that there is a new stability limit associated with each new state. But the stability limit associated with the current or post-contingency operating state is not unique, because it depends on how the limit is computed. Stability limits do exist, though they are not fixed; they change with the power system’s loading, voltages and topology; and depend on the way the operating conditions are stressed up to reaching instability. Such a dynamic nature of stability limits has led to its on-line tracking by Real-Time Stability Assessment Software (RTSAS) in the control centers [2].

 

RTSAS leads to more secure power system operation whilst operators aided by automatic control actions may handle the operating point. Even some slow developing disturbances may be handled by this process. Disturbances originated by fast developing chains of events can only be prevented by prompt automatic actions that bring the operating point back to conditions below conservative limits. RTSAS shall address transient stability, voltage stability and steady state stability [2].

 

Transient stability limits shall be understood as how much a power system or part of a power system can be loaded before the occurrence of instability. Voltage stability limits are related to the maximum active power that may be transferred to a load bus prior to a voltage collapse. Steady state stability limits can be defined as the maximum gradual load change that a power system shall withstand and remain stable [3]. Small signal oscillations shall also be monitored by automatic means at the interconnections of large power systems [4].

 

System Integrity Protection Schemes (SIPS) break chains of events that may lead to large disturbances. SIPS aim to prevent instability whenever fast chains of events take place and manual and automatic control are not capable of keeping the power system under satisfactory conditions. SIPS provide drastic automatic actions intended to bring the operating point back to conditions below conservative limits. For future networks SIPS shall become smarter by using seamless communication schemes; by being integrated into Wide Area Monitoring, Protection and Control Systems (WAMPACS); by benefitting from IEC 61850 based secondary systems and by benefitting from an efficient data and information structure [5].

 

Once DER and Smart Grids trend to be installed in larger and larger scales, power systems shall be operated, monitored, controlled and protected bearing in mind different conditions that include more uncertainties than those considered today. The electric energy industry models push towards operating power systems with smaller security margins than today. Preventing power system instability in future networks will demand better coordination and further integration between RTSAS and SIPS.

 

Classical instability countermeasures shall remain in practice, for example:

  • Angular instability: consistent policy for the application of out-of-step blocking of distance relays;
  • Voltage instability: automatic blocking of on-load tap changers.

 

 

10.2 Large loss of generation – underfrequency – (BR)

 

Frequency stability is characterized by the ability of a power system to maintain the frequency within an acceptable range after severe disturbances with cascade shutdowns of large generation blocks or loss of significant load amounts. Over or underfrequency conditions often go together with the occurrence of over or undervoltage phenomena.

Underfrequency is more likely to occur than overfrequency, despite the spinning reserve policies. Thus, underfrequency is dealt with by shedding non-priority load.

Underfrequency load shedding shall bear in mind the predominant type of generators and their capability to withstand low frequencies. For instance, in a power system with predominant hydro generation there are inherently larger underfrequency margins than in other system with predominant nuclear generation.

Underfrequency load shedding may be performed by measuring:

  • Plain underfrequency (81U);
  • Average rate of frequency decrease (∆f/∆t);
  • Instantaneous rate of frequency decrease (df/dt).

 

None of these measuring criteria is considered fully selective, so the trend is to shed load in a conservative mode. In certain applications there may be overfrequency immediately after load shedding. As an example, the ∆f/∆t scheme has intrinsic large errors, despite the application of accurate frequency and time measurements:

The ideal measuring condition is expressed by , where f1 and f2 are the limits of the frequency variation range (Hz) and T is the discrimination time (seconds). This condition is depicted in Figure 7.a.i.1:

 

Figure 7.a.i.1 – ∆f/∆t Measuring Philosophy

 

Bearing in mind that frequency measurement f1 and f2 have the same accuracy class εf and time measurement T has accuracy class εt, the maximum and minimum errors of the frequency rate variation will be expressed by:

Where and are the limits of the scheme’s warranted accuracy. Figure 7.a.i.2 shows [MGA71]the implications for a single ∆f/∆t measurement and Figure 7.a.i.3 shows the overlapping possibility when there is more than one ∆f/∆t measurement:

 

Figure 7.a.i.2 – ∆f/∆t Measurement Warranted Accuracy

 

Figure 7.a.i.3 – ∆f/∆t Measurements Risk of Overlapping

 

More accurate and selective underfrequency load shedding may be implemented by novel algorithms, such as the Generic Frequency Parameter method [6].

 

Certain DER processes, such as wind generation, cooperate in the frequency control by behaving like a negative load. The difficulties in forecasting precisely the DER generation values introduce new uncertainties to the spinning reserve quantification, implying in the need of closer coordination between spinning reserve and underfrequency load shedding. A similar process is likely to occur concerning load forecast with the advent of smart grids.

10.3 Transient stability – (BR)

 

The transient instability is characterized as the inability of generators in maintaining synchronism when subjected to a serious disturbance in the power system, which involves large excursions in the generators angles, significant variations in active and reactive power flows, voltage variations, etc. Quoting [2] on transient stability limits:

“Time domain transient stability analysis is both accurate and flexible [7] in terms of modeling detail and can handle:

  • All the known types of power system components that correspond to active injections, such as generators, loads, static VAr compensators (SVC), and FACTS devices, as well as the associated controls;
  • Any type of contingency, including three-phase and single-phase faults, as well as outage of multiple transmission and active power system components;
  • Any type of instability, such as first-swing or multi-swing, up-swing or back-swing, and plant or inter-area mode.”

Future networks are expected to have DER connected to the grid through modern inverters that may change the “signature” of the fault currents, for instance, just providing positive sequence current contribution for ground faults. Thus, both angular stability assessment tools and SIPS shall have algorithms that consider such a peculiarity. More, any adverse aspect stemming from DER and smart grids shall be contemplated when dealing with angular stability.

 

Proposition:

Unless the adverse aspects mentioned above be known by other WG Members, it would be advisable to send a query on this issue to all SC B5 and IEEE PES PSRC members.

 

In 1964, a Special Report of CIGRÉ Group 32 stated that “any network that meets the steady-state stability conditions can withstand dynamic perturbations and end in a stable operating state” [8]. The steady-state stability limit (SSSL) is a steady-state operating condition for which the power system is stable in steady-state, but for which an arbitrarily small change in any of the operating quantities in an unfavorable direction causes the power system to lose stability [9]. A transient stability limit (TSL) can be imagined, once there is no specific formula to quantify it. TSL shall be determined by transient stability simulations for each potential disturbance until the first unstable state has been identified. Once this approach is not practical, a “safe” active power grid utilization may be expressed as a fixed percentage of SSSL (security margin) and could be determined such that: for any power system state with a steady-state stability reserve higher than this value, no contingency, despite its severity, would cause transient instability [2].

 

Future networks are expected to present more volatile TSLs due to the uncertainties that are being gradually introduced by DER and smart grids. This is a major concern for dimensioning SIPS in terms of arming, disarming and triggering actions.

 

SIPS used to prevent transient instability shall provide fast response based on the accurate identification of power system contingencies [10]. Technological advances in the development of hardware and software, associated to the development of simulation techniques involved in the definitions of settings, allow the implementation of more elaborate SIPS.

 

SIPS threshold settings or reference values shall be determined so as to make the scheme more selective. A single condition of balance between selectivity and reliability certainly will not be the optimum solution for all the contingencies and respective actions that SIPS shall deal with. SIPS for the future network shall preferably be adaptive in order to allow an optimization of selectivity and reliability. Adaptive features shall be implemented under a thorough analysis of introduction of new failure modes and the corresponding recovery mechanisms [11].

 

Summing up, SIPS of the future are expected to [12]:

  • Provide:

-        Self-healing capabilities for the power system,

-        Micro-grid islanding (preventing spurious islanding) and restoring,

-        Power flow control during disturbances,

-        More accurate generation or load shedding,

-        Emergency generation reduction (no tripping);

  • Benefit from:

-        Early warning systems,

-        Monitoring of transmission assets,

-        Real-time measurement,

-        Seamless communication systems (smaller latency and larger bandwidth),

-        Improved automation technology devices and systems (IEC 61850, synchrophasor measurement, etc.),

-        Limitation of fault currents in the transmission system and from micro-grid dispersed generation;

  • Become technically:

-        Less heuristic and more holistic,

-        More predictive and then (later) more proactive,

-        More fuzzy,

-        Easier to implement thanks to protocol evolution and uniform characteristics,

-        More in tune with cyber security needs,

-        More prone to assist in post-disturbance analysis;

  • Become under the consumer viewpoint:

-        Better understood by consumers,

-        More associated to energy quality,

-        More involved with energy tariffs,

-        More subject to consumer legal aspects.

[VL72]

 

One may say that the human being is responsible for every and each failure that occurs in a power system, because the power system; its components; its planning, engineering, operation and maintenance are produced or done by human beings. Theoretically a power system could be fail-proof if there were enough funds to afford it. Once funds are limited, failures and disturbances arise as a consequence of what human beings could do with the available funding.

 

If human errors are considered, then statistics show that such errors have been involved in almost all cascading power system events.

 

 

 

 

10.4 Transient stability / angle instability / islanding – Mota / Adam / Alex[MGA73]

As discussed in previous chapters, the increased need for Wide Area Monitoring, Protection and Control Systems (WAMPACS) will arise in the future grid, for example, due to a large amount of DER connected to the power system and the introduction of Smart Grids.

 

Since 1970s, a number of WAMPACS have been operated in Japan ( ([38]) ([39]) ([40]) ([41]) ([42]) ([43]) ([44]) ([45]) ([46])). The application includes the last resort for the transient stability and the voltage stability as well as the islanding protection. The service experiences of these WAMPACS have been satisfactory as successful operations have been reported in literature  ([47]).

 

Figure 7.b.i.1 shows an conceptual diagram of one of WAMPACS in Japan which operates based on the on-line transient stability assessment. The WAMPACS is installed near a generation center and shed minimum generators required to keep stable operation of the rest of the system. For the on-line transient stability assessment, the WAMPACS solves the swing equation by modeling the system as two equivalent machines, one for the generation center and the other for the rest of the system. The WAMPACS calculates the future phase angle difference between these two machines, and commands to shed generators when the future phase angle difference exceeds a setting value.

 

In this calculation, the priority of generators to be shed is set in advance. The number of generators to be shed is minimized by the repeated calculation. The WAMPACS calculates the future phase difference repeatedly, increasing the number of generator to be shed in each repetition, and find the optimum generator shedding which limits the future phase angle difference within the setting value.

 

Such WAMPACS will be benefitted from the current trend for the standardization, particularly in communication, time synchronisation, synchrophasors, and security. A WAMPACS based on the international standards, such as IEC 61850, IEC 61970, IEEE C37.118, and IEEE 1588 V2, has already been built for a demonstration ([48]). In the demonstration WAMPACS, IEDs/PMUs and PDCs are used instead of legacy RTUs. Satisfactory results in the evaluation test suggest that WAMPACS in the future grid will be based on the international standards, providing interoperability between IEDs of different manufacturers.

 

Figure 7.b.i.1 – Conceptual diagram of the example WAMPACS

 

10.5 Voltage Instability Stability Peter[MGA74]

See section about voltage instability below[MGA75]

 

10.6 Causes of cascading power system events[i] [1][VL76]

 

A cascading power system event – a system-wide disturbance – is a continuity of the chain of events that originate a disturbance. The cascade effect direct or indirectly involves human failures. Cascading power system events are generally originated from phenomena such as angular stability, voltage stability, overloads, multiple contingencies, etc.

Angular stability[ii] [iii][2 and 3] is a key factor for power system secure operation. Assuming that spinning reserve is adequate, single or multiple contingencies in the transmission system are expected to be controlled by regulators and stabilizers, causing only dynamic oscillations among generating units and among power system areas, preserving synchronism all along the power system. If one or more TLs are tripped during an intense and long lasting stable oscillation, synchronism may be lost and a cascading effect may be started. Some contingencies, typically multiple ones, may be too severe to be contained by regulators and stabilizers, implying in immediate loss of synchronism.

Even providing excellent out-of-step tripping at strategically selected points, a disturbance of some extent may be configured, once islands are created and each one will have to cope with issues as:

  • Unbalance between generation and load, as well as the corresponding frequency conditions, having to automatically shed the excessive amount of generation or load;
  • Island grid topology to avoid internal angle stability problems;
  • Island grid transmission capacity to avoid internal angle stability problems and overloads.

 

Voltage stability is defined by the System Dynamic Performance Subcommittee of the IEEE Power System Engineering Committee[iv] [4] as the “ability of a system to maintain voltage such that when load admittance is increased, load power will increase, so that both power and voltage are controllable”. Also, voltage collapse is defined as being the process by which voltage instability leads to a very low voltage profile in a significant part of the system.

It is accepted that this instability is caused by the load characteristics. The risk of voltage instability increases as the transmission system becomes more heavily loaded. One typical scenario for this kind of instability starts with a high system loading, followed by a relay action due to a fault, a line overload or an excitation limit reached.

 

Outage of one or more power system elements due to overload may cause overload in other elements in the system. If the overload is not alleviated in time, the process of power system cascading may start, leading to power system separation.

Multiple contingencies are typically caused by[v] [vi][5 and 6]:

  • Evolution of a localized fault by trips initiated from conventional backup protection, protection false trips and those due to hidden failures from protection functions;
  • Sequential faults;
  • Severe weather conditions;
  • Geomagnetic induced currents;
  • Natural disasters such as earthquakes;
  • Poor maintenance of primary system equipment.

[VL77]

Failures or misoperations in protection systems have a significant influence in the overall process of system-wide disturbances:

  • The application of conventional backup protection functions is not recommended along the power system main grid, precisely from where system-wide disturbances are originated. Conventional backup protection shall not be applied to the main grid components because it is based on grading protection philosophy. Main grid components are typically protected by redundant protective schemes based on unit protection philosophy plus selective breaker failure schemes;
  • Protection false trips are the consequence of defects not detected by self-testing or self-supervising sub-functions of a protective scheme or IED. They do not constitute a typical cause of large disturbances, once failures that lead to an immediate misoperation during normal power system states can be corrected right away;
  • Hidden failures are particularly dangerous. They remain dormant or hidden until the power system conditions reach some uncommon state and only then these failures show up by tripping circuit breakers. One of the power system abnormal conditions may trigger the hidden failures to cause protection function misoperations which worsen the situation, eventually leading to system-wide disturbances.

[VL78]

 

10.6.1 Prevention against cascading power system events1[VL79]

Before dealing with the approaches for prevention of each kind of mentioned phenomenon, it is interesting to remind the mechanics behind essential on-line conditions [6], as per Figure 7.b.ii.1:

 

Description: fig

Figure 7.b.ii.1 – Power System States: Fink and Carlsen Diagram

 

Ideally, regulators, system stabilizers and system component protection would be enough to assure high electric reliability. Nevertheless, the trip of some specific system components can stress the power system to conditions beyond the capabilities of regulators and system stabilizers. For many of these cases WACS will be necessary to return the operating state to a safe condition. Also, there are events, generally cases of multiple contingencies, which will require stricter – and more painful – actions to avoid the power system collapse.

 

Indeed, the prevention starts in power system expansion planning by, among other issues, avoiding:

  • Agglomeration of too many transmission lines in the same right-of-way;
  • Erection or expansion of excessively large substations;
  • Crossing of aerial new lines over existing transmission trunks;
  • Indiscriminate short circuit level growth [7];
  • Poor busbar arrangements.

 

Many other issues could be raised to illustrate why the power systems may be subject to temporary or nearly permanent weaknesses. One has to admit that immunity against system-wide disturbances in an environment of an infinite number of possibilities of unexpected conditions is, in fact, impossible. SIPS are the path to assure that these possibilities will be minimized.

 

The prevention against angular instability may be implemented by fast logical detection of dangerous topologies, by out of step protection or by predictive protection schemes.

 

Logical detection of dangerous topologies is usually implemented by circuit breaker and switch auxiliary contacts supervision. Logical detection of events is a fast and inherently redundant criterion to cope with a high degree of multiple contingencies, by automatically shedding generation to tune the power flow through the remaining TLs and transformers according to the power transfer capacity of the resultant topology.

 

Logical detection schemes depend on supervision for discrepancy of switch secondary contacts, which allows the designer to privilege SIPS dependability or security according to the influence of the temporary information loss. Memorizing the last correct state of switches is another way of implementing logical decisions in this case. Logical detection criterion may be implemented by SAS or even by a Remote Terminal Units (RTU). Inherent redundancy comes from the use of data from two different substations in case of circuit tripping or from two voltage levels in case of transformer tripping.

 

Out-of-step protection [5], as it is applied to generators and systems, has the objective to eliminate the possibility of damage to generators as a result of an out-of-step condition. In case the power system separation is imminent, it should take place along boundaries, which will form islands with matching load and generation. Distance relays are often used to provide an out-of-step protection function, whereby they are called upon to provide blocking or tripping signals upon detecting an out-of-step condition.

 

The most common predictive scheme [5] to prevent loss of synchronism is the Equal-Area Criterion and its variations. This method assumes that the power system behaves like a two-machine model where one area oscillates against the rest of the system. Whenever the underlying assumption holds true, the method has potential for fast detection.

 

Voltage instability can be alleviated by a combination of the following remedial measures means: adding reactive compensation near load centers, strengthening the transmission lines, varying the operating conditions such as voltage profile and generation dispatch, coordinating relays and controls, and load shedding. Most utilities rely on planning and operation studies to guard against voltage instability. Many utilities utilize local voltage measurements in order to achieve load shedding as a measure against incipient voltage instability [5 and 12].

 

Overloads may have very distinct origins, since the trip of one in two transformers that supply a radial load larger than the capability of the remaining transformer, up to transient conditions imposed to a whole trunk when the power system is at the edge of a blackout. In this last case if the imbalance cannot be handled by the generators, load or generation shedding is necessary. A quick, simple, and reliable way to re-establish active power balance is to shed load by underfrequency relays. There are a large variety of practices in designing load shedding schemes based on the characteristics of a particular system and the utility practices [5].

 

Summing up, SIPS are the way through which cascading power system events shall be dealt with. But even SIPS are subject to human errors…

Cascading power system events (Human in the loop) in future networks

 

For future networks, the prevention of cascading events shall stress two main issues:

  • Safety margins to cope with uncertainties;
  • Real-time automatic support for all human activities.

 

International benchmarks shall be produced by collective work (CIGRÉ and IEEE) about the uncertainties introduced by DER and smart grids. From these, safety margins and respective application methodologies shall be inferred as the most useful deliverables of joint activities on this subject.

 

Angular and voltage stability assessment systems are perfect examples of real-time automatic support for human activities, though restricted to control center operators. New, more sophisticated and more user-friendly systems and tools for aiding operators are welcome, for instance, like Brazilian Organon system [3 and 9].

 

Concerning secondary systems, IEC 61850 inherently provides real-time automatic support for its users:

  • Manufacturing Message Specification (MMS):

-        MMS protocol based information is available to flow upward along the hierarchy, for SCADA and EMS functionalities,

-        Likewise, it may flow downward, being typically addressed for commands,

-        The Model Implementation Conformance Statement (MICS) document of each IED presents the available data objects that help organizing DIS;

  • SAS Engineering Tools [10 and 11]:

-        SCL is used also to describe all data needed to define system parameters for a single IED,

-        This includes the binding of the IED and its functions to the substation itself –from its single line diagram to the communication system,

-        The Substation Configuration Language is based on UML and XML, which:

  • Offers a method for putting structured data into a text file,
  • Looks a bit like HTML,
  • Is machine readable, but human intelligible,
  • Comprises of a family of technologies,
  • Is verbose,
  • Is license free, platform-independent and well supported,

-        SCL components:

  • Substation section describes the substation single line diagram, and its binding to logical nodes as well as the placement of logical nodes onto IEDs, thus also the binding of IEDs to substation parts and substation devices is defined,
  • Communication section describes the communication connections between IEDs in terms of connecting communication links,
  • IED section describes the capabilities (configuration) of one or more IEDs, and the binding to logical nodes on other IEDs,
  • LNType section defines which data objects are actually contained within the logical node instances defined for the IEDs,

-        SCL files:

  • Data exchange from a system specification tool to the system configuration tool describes the single line diagram of the substation and the required logical nodes, the file extension shall be .SSD for System Specification Description,
  • Data exchange from the IED configuration tool to the system configuration tool, describes the capabilities of an IED, the file extension shall be .ICD for IED Capability Description,
  • Data exchange from the system configuration tool to IED configuration tools, contains all IEDs, communication configuration and substation description sections, the file extension shall be .SCD for Substation Configuration Description,
  • Data exchange from the IED configuration tool to the IED describes an instantiated IED within a project, the communication section contains the current address of the IED, the substation section related to this IED may be present and have name values assigned according to project specific names, it is an SCD file, possibly stripped down to what the concerned IED shall know, the file extension shall be .CID for Configured IED Description,

-        Engineering efficiency improvement will be possible by extending the use of the SCD files for the following substation automation systems engineering tasks:

  • Automatic creation of the graphical user interface from the SCD file, including the different screen layouts,
  • Automatic mapping of the different measurements and status information from the IEDs to the substation HMI,
  • Automatic configuration of the IED or substation protection and control system testing process,
  • Automatic substation event analysis.

 

IEC 61850 eases the implementation of a PAC Data and Information Structure (DIS) from the process up to national control center levels, in other words, a multi-layer hierarchical country-wide power system control structure. DIS will provide the creation of high added value information from the lowest possible level of the operational hierarchy, alleviating communication media and offering better flexibility for implementing protection, automation and control macro-functions, thus improving their performances. Figure 7.b.iii.1 summarizes the main levels of Brazilian DIS and their relative needs for data and information:[VL80]

 

Figure 7.b.iii.1 – Data and Information Relative Requirements

 

  • Future distribution system (33% DER) fault

10.7 Future Distribution System

10.7.1 Grid-connected

Because DER-units are contributing to ancillary services (refer to 5.3.4) they contribute to the short-circuit-current while the network protection operates, also for disturbances in the distribution grid. From this principle reason arise impacts for the network protection and reasons for loss of reliability of protection systems in distribution grids. Table 5Table 3 summarizes ten aspects of impacts with their inherent reasons for loss of reliability of protection.

 

Table 53. Summary of the problem area DG and system protection

A1

DG units now contributing to voltage support, ride through, and frequency regulation

1. Additional short-circuit current contribution from DER units during the tripping time of protection devices

A2

Intermediate infeeds

1. Measuring error of the measuring element

A3

Bidirectional powerflow

  • Sympathetic tripping of adjacent relays
  • Auto Voltage regulation issues

A4

Increased ampacity during normal and abnormal power system conditions

1. Discrimination of normal and abnormal power system conditions

2. Cables in parallel operation

3. Fault current limiter

A5

Unintentional islanding – designed to island / non-desirable island

1. Undetectable island operation

A6

Controlled islanding – Transmission/Distribution/Microgrids

  • Abnormal (faulted) conditions during controlled islanding
  • Microgrid operation (frequency, voltage, load control)

A7

Voltage applied from the remote end during system protection trippings

1. Automatic reclosure, increased dead-times

2. Transient overvoltages, transient recovery voltage during out-of-step switching of circuit-breakers

A8

Coordination of protection devices to ensure selectivity

Coordination with protection of the DG device

 

B1

Different and unusual short-circuit behavior of different power generation technologies

1. Voltage sources (induction principle) Current sources (self-commutated converter)

2. Limited short-circuit current contribution

3. Short-circuit current differs from 50Hz / 60Hz – quantify this difference??? - Timo

B2

Transients of small Synchronous power generation units

1. Transient stability, small coefficient of inertia

2. Moving measurands (current and voltage phasor)

3. Power swings

B3

Variable generation infeeds (windfarms, solar)

1. Variance between load and generation

2. Variation of short-circuit conditions

 

10.8 Island grid – all inverter based generation – Peter

Distribution level power swing

 

10.9 Long duration faults (ride-through & VAR support issues) – Nabil

  • Overloads (non-fault)

Use case for long duration faults:

1.- Define wind penetration level as a percentage (%) of grid rated power

2.- Scenarios of wind power generation conditions: full load (100% rated power), partial load (10 ~30% of rated power)

3.- Events within the simulation time:

Fig. 1 shows the events simulated for a long duration fault:

Fig. 1.  Events simulated in a long duration fault with a distributed generation network model.

a) A flat simulation time should be performed to verify that distributed generation network model get stabilize to its rated values.

b) Event 1: Once consumed flat simulation time, a line fault is applied. Voltage dips varies from 0% up to 80% of rated voltage while fault length can be considered as the time of a zone 2 temporization (150 ~ 250 ms).

c) Event 2: Following the fault isolation, a disconnection of 6 ~ 10 % of the generation simulated by the grid will get disconnected. This will allow frequency to drop down and to move the frequency control loops implemented in the modeled network

Voltage and frequency variations expected for simulation event 1 and 2 are depicted in Fig. 2

(a)

(b)

Fig. 2.  (a) Voltage and (b) frequency variations expected for Event 1 fault simulation.

 

During fault length and ride-through fault, the DER will response according to grid codes requirements [1]. Notice that frequency is stabilize to a lower magnitude.

d) Event 3: Load shedding is implemented to allow frequency to recover to 50 Hz, i.e. a successive disconnection of load will be performed until the network get once again to a balance point that is generation power fits the load demand. Fig. 3 shows the complete event simulation of the long duration fault. Notice that frequency recovers to 50 Hz after load shedding.

  • Fig. 3.  Frequency variation for the long duration fault event.

 

10.10                      Automation based on Asset condition – Peter/Alex (switching to protect assets)
 

Issues, Potential Solutions, Coordination and interaction of protection and automation associated with Future Networks

  • Dynamic protection setting coordination and verification- Rodney

 

In order to prevent  major disturbance  leading to power system collapse ,it is necessary to isolate part of the network that can survive in isolation with its own generation and Load. This is termed as Island..   The islanding scheme pre-supposes that the integrity of the system cannot be maintained in spite of various protection and control interventions.  Instead of allowing the system to disintegrate by the tripping of generators and transmission lines as the disturbance develops, the islanding scheme itself sectionalises the whole system into sustainable small systems each consisting of a group of generating stations and a group of load that can be supplied by these generating units.  In effect each group becomes a sustainable island.  Such a scheme has to be carefully planned and engineered. Protection schemes detect certain system parameters based on which isolation strategy is implemented or islanded condition is detected

The methods of detection, protection and re-connection used in islanding scheme will differs based on several system parameters and characteristics of island. The islanded grid could be :

 

 

Region : A complete region may get separated from rest of the interconnected Grid.

Area : An area could be a section of regional grid or it could be a dedicated netowrk for captive load (industrial power system) or a mirco grid.

Station : A station comprising of several generating units

Generating unit : a generating unit isolates itself and survives on its houseload, ready for re-connection.

Detection : There are several methods that are used for detection. Any one of the method in it self or in combination with other parameter is used to initiate the islanding.

Some common conditions are

- Rate of change of frequency

- Under frequency

- Direction of Power flow

- Under voltage / over voltage

- Phase angles / vector shift

It s important to select carefully a point of isolation or points of isolation. It is usually a connection and reconnection point with rest of the grid. Direction of powerflow at these points is critical. If there is export of power, island will be generation surplus leading to over frequency. If it is import point, the island will have deficit and frequency would drop after islanding. It is of great importance that the detection is done accurately and fast sothat necessary controls can be activated.

For Distributed Energy Resources (DER) Island operation occurs if one or more distributed energy resources continues to energize a part of the grid after the connection to the rest of the system has been lost. Island operation can be either intentional or unintentional. In the first case, the islanding has been planned in advance and the system and equipment has been designed to cope with the situation. The DER is then well suited to control voltage and frequency in the islanded grid. Intentional islands often exist in industrial plants where the process has surplus energy that can be used to produce electricity.

 

In case of DER, Generally used methods of detection are broadly classified as

1. Passive Methods

2. Active Methods.

Passive methods :- Locally available quantities such as voltage, curent , frequency,phase angles etc are used.

 

Active Methods :

 

 

Protection of island :

Key difference between interconnected operaiton with Grid and Islanded operation lies in the fact that the system intertia changes significantly. In islanded grid small change in generation or load can change the frequency significantly ,depending on capacity of Island. Also the fault current levels will change drastically there by protection coordination considered earlier may not be valid. Hence new set of protection settings will have to be adopted and automatic control actions are required to stabilise the islanded grid. Few such schemes are :

- Under frequency based load trimming

- Over fequency based auto load restoration

- capacitor / reactor bank switching or SVC controls

- special protection schemes for low fault level operation

- Supplimentary controls for generation units to react to changes and stabilizing actions

- Synchronizing scheme to reconnect the island

 

 

 

  • Distribution

23/10/2012
[VL81]Classification of automation functions is provided in IEC 61850 standard using the model of logical nodes. A wide range o flogical node types is provided in the standard to define data model for the primary equipment and majorautomation functions.  In particular, the future work item IEC 61850-7-500 will demonstrate use of logical nodes to model functions of a substation automation system.



  •  
  • Local, regional and wide area automation for Distributed Energy Resources (DG / RES/Energy storage/Loads) (supervision, adaptation to network topology and constraints) - ??
  • Automation functions to monitor and adjust transfer capacity (watts and vars) on different network levels - ??
  • Local or regional automation functions related to the stability of the generating and higher-voltage resources - ??
  • Impact of FACTS on line protection – Sankara / Mark
  • Protection of Inverter Based Transformers – Peter’s student

10.11                      Adaptive protection responding to network changes resulting from automation – Volker / Andrei / Tianshu

10.11.1                    Case : Skegness-Boston DLR Scheme [NE of England]

Wind farmsor other RES are sometimes located at the extremes of the distribution system or on tie-lines of the subtransmission system. These lines may not be rated to carry the full output of the wind farm in all circumstances. Often a line has been designed originally to supply a relatively small load, and the installation of e.g. new wind generation may cause a large reverse power flow, causing the standard winter and summer line ratings to be exceeded. Rather than applying fixed summer and winter line ratings, adaptive overload protection may be used.

In this approach, the maximum load current thresholds are calculated dynamically taking into account parameters like windspeed and –direction, air temperature, solar radiation and an estimated or measured conductor temperatures. Estimation can be based on previously recorded load currents applying a thermal model of the line.

The diffuculty of this scheme is to determine the proper algorithm calculating load thresholds from all these variables. It is clear that ponderation of most parameters has to take into account characteristics of the line like orientation and exposure to wind of its different segments.

The analysis conducted for the 132 kv double-circuit line Skegness-Boston in the UK shows that the weather parameters having a significant impact on the line rating. In the order from lowest to highest, this impact is ([49])

  • solar radiation,
  • ambient temperature (max +5%),
  • wind speed (max +92%),
  • wind speed + wind angle (132%).

 

10.11.2                    Case : Adative distance protection using network topology

Settings of distance protection have to take into account "worst cases" when determing the possible over- or underreach related to strong or weak sources at each terminal of the protected line. In some cases, this leads to prolonged fault elimination times in some cases, which have to be accepted in order to avoid unselective tripping for other configuration.

Online knowledge of the network topology and of connected sources allows to establish optimized thresholds for each moment, for example

  • by determing the equivalent source impedance at each line end.
  • by knowing whether the circuits of a double-circuit-line are connected to the same busbar at the two line terminals.
  • by the assement of the lumped equivalent of DER injection at a given point in the network.

These parameters enable the automatic selection of predermined setting groups or an adaptive on-line calculation of the different thresholds.

 

 

11.           Transition to the Future Grid

 

Conclusion to be written

12.           Bibliography

 

13.           Bibliographie

[01]. JWG B5/B4.25 Impact of HVDC Stations on Protection of AC systems. CIGRE.

[03]. www.twenties-project.eu.

[04]. C. Ribeiro, E. Fontana and J.M. Ordacgi “Smart Grid in Brazilian Transmission System - Views from Primary and Secondary Systems,”. Paper P005, Protection, Automation & Control World Conference 2010, Dublin, Ireland, June 2010.

[05]. V. Vyatkin, G. Zhabelova, N. Higgins, M. Ulieru, K. Schwarz, N. Nair, “Standards-enabled Smart Grid for the Future EnergyWeb”. The 1st IEEE Conference on Innovative Smart Grid Technologies, Gaithersburg, MD, January 19-21, 2010.

[06]. S. Grobbelaar and M. Ulieru, “Complex Networks as Control Paradigm for Complex Systems,”. 1st IEEE International Workshop on eNetworks Cyberengineering, IEEE, 2007.

[07]. Eko Adhi Setiawan, “Concept and Controllability of Virtual Power Plant”. Thesis for acquiring the academic degree of Doktor der Ingenieurwissenschaften in the University of Kassel, Germany, 27 April 2007.

[08]. Onisova O., Nudelman G., Navolochnyi A. Relay Protection and Automation in Power Systems with Distributed Generation / Relay Protection and Automation for Electric Power Systems. 2012 International Exhibition and the 21st Conference, Moscow, 29-31 May 2012.

[09]. P. Sankarakumar, K. Venkataraman (Alstom Grid)- Use of Transmission Protection Technology In Distribution Relays. DPSP 2012 - Birmingham.

[10]. E. Cloet, J-L. Lilien and P. Ferrieres, “Experiences of the Belgian and French TSOs using the “Ampacimon” real-time dynamic rating system"; Paper C2_106_2010. CIGRE 2010 Paris.

[11]. F.M. Silva, “Response to Special Report Question 1.3,” Preferential Subject 1, Plenary Session of CIGRÉ SC C2. CIGRE 2010 Paris.

[12]. Dufour,J. Feasibility study for the protection of long underground lines – Paper 209. CIGRE B5 Colloquium 2011, Lausanne.

[13]. Y. Vernay; J. Dufour - “Electrical studies performed to insert long AC cables in the French grid – first conclusion”. JiCable 2011 Versailles - France.

[14]. Kumpulainen, Pettissalo, Trygg, Malmberg, Loukkalahti, Hyvärinen - "Monitoring and Control of Distribution Network by Smart MV/LV Substations”. IEEE-PES ISGT Europe 2011.

[15]. JWG B5/B4.25 - CIGRE TB 484 - "Impact of HVDC Stations on Protection".

[16]. (n.d.). N. Higgins, V. Vyatkin, N. Nair and K. Schwarz,"Intelligent Decentralised Power Distribution Automation with IEC 61850, IEC 61499 and Holonic Control". IEEE Transactions on Systems, Man and Cybernetics, Part C, 41(1) - 2011 , pp. pp.81-92.

[17]. (n.d.). G. Zhabelova, V. Vyatkin, “Multiagent Smart Grid Automation Architecture Based on IEC 61850/61499 Intelligent Logical Nodes”. IEEE Transactions on Industrial Electronics, 2012, 59(5), , pp. Page(s): 2351 - 2362.

[18]. (n.d.). Dugan, R.C. & McDermott, T.E. (2002). Distributed Generation. IEEE Industry Applications Magazine, Vol. 8, No. 2, (March-April 2002), , pp. pp. 19-25, 1077-2618.

[19]. Yuping Lu; Lidan Hua; Jian Wu; Gang Wu & Guangting Xu (2007). A Study on Effect of Dispersed Generator Capacity on Power System Protection. Proceedings of IEEE Power Engineering Society General Meeting,2007, pp. 1-6, 1-4244-1296-X June 2007. Tampa,FL.

[2]. T.K Vrana, O. B. Fosso - Technical aspects of the North Sea Super Grid . CIGRE - ELECTRA No. 258.

[20]. M.Paz Comech, Miguel García-Gracia, Samuel Borroy, M.Teresa Villén - "Protection in distributed generation".

[21]. (n.d.). Morren, J., de Haan, S.W.H. (2007)-" Short-Circuit current of wind turbines with doubly fed induction generator". IEEE Trans. On Energy convers, vol. 22, no. 1, march 2007 .

[22]. CIGRE Technical Report 339 “Guideline on the impacts of fault current limiting devices on protection systems”, WG A3.16, 2008. CIGRE.

[23]. Jaeger J. et al.: "Interaction between FCL on Protection – a glance at the final report of CIGRE WG-A3.16". IEEE PES General Meeting, 07.2008. Pittsburgh, USA.

[24]. aeger J.- Impacts of fault current limiting devices on protection systems". CIGRE Session 2008, SC A3, 08.2008. Paris.

[25]. CIGRE WG A3.16: “Fault Current Limiters - Application, Principles and Experience”, paper 205. CIGRE SC A3&B3 Joint Colloquium "Present and Future of High Voltage Equipment and Substation Technologies", Tokyo 26-27 September 2005,, (pp. pages 187-192.).

[26]. ”Analysis of Protection Relaying Requirements for Solid State Current Limiter Applications”. EPRI, Palo Alto, CA: 2001.1001969.

[27]. (n.d.). H. Kameda, H. Taniguchi - “Setting Method of Specific Parameters of a Superconducting Fault Current Limiter Considering the Operation of Power System Protection - Resistance-type and Rectifier-type SFCLs in Overhead Transmission Systems”. IEEE Transactions on Applied Superconductivity, Vol. 9, No.2 , 1355-1360.

[28]. L. G. Kenneth P. Birman, Robbert van Renesse, "Running Smart Grid Control Software on Cloud Computing Architectures". Workshop on Computational Needs for the Next Generation Electric Grid, Cornell University, Ithaca, NY, 2011.

[29]. CIGRE WG 34.02 TB 49, “Adaptive Protection and Control”, August 1995. CIGRE.

[30]. Yi Hu, Darnir Novosel, Murari M. Saha, Volker Leitloff :" Improving Parallel Line Distance Protection with Adaptive Techniques". IEEE PES Winter Meeting, 2000, 23-27 Jan 2000.

[31]. M.J.Damborg, M. Kim, J. Huang, S.S. Venkata, A.G. Phadke "Adaptive Protection as Preventive Emergency Control". IEEE PES Summer Meeting 2000.

[32]. (n.d.). S.H. Horowitz, A.G. Phadke, J.S. Thorp, 1988, “Adaptive Transmission System Relaying”. IEEE Trans. Power Delivery, vol. 3, no 4 , pp. 1436-1445.

[32]. (n.d.). S.H. Horowitz, A.G. Phadke, J.S. Thorp, 1988, “Adaptive Transmission System Relaying”. IEEE Trans. Power Delivery, vol. 3, no 4 , 1436-1445.

[34]. (n.d.). "Optical Current Transducers For Power Systems: A Review”. IEEE Transactions on Power Delivery, Vol. 9, NO. 4, October 1994 .

[35]. IEEE Standard PC37.235-2007, “Guide for the Application of Rogowski Coils used for Protective Relaying Purposes”.

[36]. (n.d.). “230 kV Optical Voltage Transducers Using Multiple Electric Field Sensors”. IEEE Transactions On Power Delivery, Vol. 17, No. 2 .

[37]. Paper 23-208 ; “An Electro Optic Voltage Transducer”, . CIGRÉ 1996. Paris.

[38]. Y. Ohura, et al.paper 202-6 - “Microprocessor-based stabilizing control equipment for survival of isolated mid-city power system”. IEEE, 1986 Winter Meeting.

[39]. M. Takahashi, et al. paper 093-8- “Fast generation shedding equipment based on the observation of swings of generators”. IEEE, 1986 Winter Meeting.

[40]. Y. Ohura, et al. paper 125-5 “A predictive out-of-step protection system based on observation of the phase difference between substations”. PWRD, IEEE, 1990 Winter Meeting.

[41]. (n.d.). Y. Morioka, et al. “System separation equipment to minimize power system instability using generator's angular-velocity measurements”. IEEE Transactions on Power Delivery, volume 8, No.3, July 1993 , pp. 941 – 947.

[42]. K. Akiyama, et al. “Development of a new out-of-step separation system to balance supply and demands after power system separation”. IEE DPSP, 1994.

[43]. T. Inoue, et al. “Development plan for wide-area power network blackout prevention system based on online stability calculation”. CIGRE SC-39 colloquium, 1993.

[44]. (n.d.). K. Matsuzawa, et al. "Stabilizing control, system preventing loss of synchronism from extension and its actual operating experience". IEEE Trans. Power Systems, vol.10, Issue. 3, Aug. 1995 , pp. 1606 - 1613.

[45]. M. Tsukada, et al, "New Stabilizing Protection Systems with an Adaptive Control Approach". CIGRE SC34, Stockholm Colloquium, 1995.

[46]. S. Agematsu, et al. “Islanding protection system with active and reactive power balancing control for Tokyo metropolitan power system and actual operational experiences”. IEE DPSP2001.

[47]. T. Ohno, et al. “Islanding protection system based on synchronized phasor measurements and its operational experiences”. IEEE PES General Meeting, 2008.

[48]. S. Matsumoto, et al. “Wide-Area Situational Awareness (WASA) System based upon International Standards”. 11th International Conference on Developments in Power System Protection (DPSP), 2012.

[49]. Ponniah Sankarakumar- response to PS3-Q12. CIGRE B5 colloquium 2011. Lausanne .

[50] http://www.decc.gov.uk/en/content/cms/meeting_energy/Renewable_ener/feedin_tariff/feedin_tariff.aspx

 

 

 

 

 

 

 

 

 


14.           Appendix 1 – Définitions

SPS – Special Protection Scheme : Within North America, NERC defines a Special Protection System (SPS) as an automatic protection system designed to detect abnormal or predetermined system conditions, and take corrective actions other than and/or in addition to the isolation of faulted components to maintain system reliability. Such action may include changes in demand, generation (MW and Mvar), or system configuration to maintain system stability, acceptable voltage, or power flows. A NERC defined SPS does not include (a) underfrequency or undervoltage load shedding or (b) fault conditions that must be isolated or (c) out-of-step relaying (not designed as an integral part of an SPS).

SIPS -  System Integrity Protection Schemes :  The SIPS are installed to protect the integrity of the power system or its strategic portions. A SIPS is applied to the overall power system or a strategic part of it in order to preserve system stability, maintain overall system connectivity, and/or to avoid serious equipment damage during major events. Therefore, the SIPS may require multiple detection and actuation devices spread over a wide area and utilize communication facilities.                           
The SIPS encompasses SPS, Remedial Action Schemes (RAS), as well as additional schemes such as, but not limited to, Underfrequency (UF), undervoltage (UV), out-of-step (OOS), etc. These additional schemes are excluded from the conventional North American definition of SPS and RAS. A conventional protection scheme is dedicated to a specific piece of equipment (line, transformer, generator, bus bar, etc.), whereas a SIPS is applied to the overall power system or a strategic part of it. Therefore, SIPS may require multiple detection and actuation devices and communication facilities. The scheme architecture can be described by the physical location of the sensing, decision making, and control devices that make up the scheme and the extent of impact the SIPS has on the electrical system.

[VL82]

 

 


15.           Appendix 2  Definitions of Wind Turbine Generator Types

 

Introduction[VL83] Type 1 and Type 2 utilizes conventional induction generators. They are essentially a constant speed system with the slight fluctuation of speed for the duration of change of load. Type-1 has a squirrel cage induction generator however; wound rotor induction generators for Type-2 have some speed control.

[MGA84]

Fig. 2. Type-1 and Type-2 WTGs

is One of the most widely used speed WTG is the Type-3. This WTG arrangement is shown in Fig.  3. This is wound rotor machine, fed through a series back to back frequency converters; having a rating of 30% of the maximum stator power rating. The total power is the arithmetic summation of powers from stator and rotor. The DFIG operates either in super synchronous, synchronous or sub-synchronous modes. Power is injected from the rotor, through the converter, into the system when the DFIG operates at super-synchronous speed. The real power is absorbed through the converter from the system by the rotor when the DFIG operates at sub-synchronous speeds. There is no power exchange through rotor at synchronous speed as voltage at the rotor is essentially DC. Reactive power is supplied to the system through d-axis excitation control on rotor in most cases. The convertor could also be used as STATCOM for dynamic reactive compensation even when the turbine generator is not operational. In earlier designs the Type-3 has been more sensitive to disturbances and would detach from the network quicker than traditional synchronous generators. The safety of DC convertors was the main cause. Now Type-3 machines employ more controls than Type-1 and 2. One popular solution for the above mentioned issuethe crowbar protection. This technique short circuit the rotor side converter with or without additional resistance and leave the DFIG as standard induction generator during the disturbance or fault and bring the convertor back after pre-defined time period. More advanced controls are being now offered by turbine manufacturers to protect rotor and control active and reactive power simultaneously.

 

Fig. 3. Type-3 Wind Turbine Generator

 

In Type-4 the wind generator is decoupled through full back to back convertor. These could be conventional generators, dc field or permanent magnet generators. The generator spins at any available rotational speed through direct coupling to the turbine. The frequency may not be 50/60 Hz at the generator end but electrical power is converted to the required grid frequency through a back-to-back converter, thus giving generator speed a wide range because of full frequency convertors. The arrangements of Type-4 are shown in Fig. 4. The grid side convertor has the ability to independently control real and reactive power to improve ride-through ability, reactive power control and voltage regulation of the electrical generator.

 

Fig. 4. Type-4 Wind Turbine Generator

 

Type-5 WTG arrangement is based on a gear box technology converting the variable wind speed to a fixed shaft speed for synchronous generator as shown in Fig. 5. The synchronous generator then supplies the electricity at grid frequency. Currently, these turbines are smaller and not mainly used but give much better advantages during integration. FRT capability of this type is as good as conventional generators. 

 

Fig. 5. Type-5 Wind Turbine Generator

 


16.           Appendix 3 – Low Voltage Ride Through Capability

 

Introduction[VL85]

The increasing amount of wind generation to existing transmission system has prompted the need for tougher grid code requirements to maintain the security of supply. The technological variations in wind generator technology available for large scale wind farms raises concerns around their ability to support the grid during system events. The abnormal behaviour of a large scale wind generator for the period of a fault may affect system stability.   One of the grid code requirements is a Fault Ride-Through (FRT) criterion. FRT is further classified into Low Voltage Ride-Through (LVRT) or High Voltage Ride-Through (HVRT) and it is the generators ability not to disconnect from system for specific fault events. Some grid operators require all newly commissioned wind generators to meet with minimum ride-through requirements. By definition, FRT requirements bind generators to operate between allowable voltage limits and stay connected to the grid within the same voltage envelope to maintain the security of supply. Normally, these FRT requirements become the part of grid code requirements established by the Transmission System Operator (TSO).

Grid operators define a ‘ride-through’ profile in order to escape circumstances where large generators are detached for grid faults. Some system operators not only require participation from generators during the fault but also post fault period towards voltage stabilization and system recovery Most traditional synchronous type generators may have the capability to fulfil ride-through; however, wind generator technology may not be generalized at this stage as it varies in capability conditional to the Wind Turbine Generator (WTG) technology employed. FRT criterion has already been developed and in practice in many countries which is summarized in TABLE I

[21] Tsili, M. and S. Papathanassiou, A review of grid code technical requirements for wind farms. Renewable Power Generation, IET, 2009. 3(3): p. 308-332.

TABLE I.               VRT Criteria in Practice [21]

Grid code

Fault duration (ms)

Fault duration (cycles)

Min voltage level

(% of Vnom)

Voltage restoration (s)