Article - Issue 53, December 2012

Balancing the energy network

Phil Lawton

Download the article (863 KB)

The National Balancing Engineer in the Electricity Transmission Control Room has to manage the system frequency at 50 Hz while keeping power flows within limits

The National Balancing Engineer in the Electricity Transmission Control Room has to manage the system frequency at 50 Hz while keeping power flows within limits

Come 2020, renewable sources are to provide up to 15% of the UK’s total energy needs. The way that energy networks operate will be different as many more coal-fired power stations have to shut down. Phil Lawton, Grid Operations 2020 Manager at National Grid, looks at how his organisation will tackle the upcoming challenges and keep the energy flowing.

National Grid connects people to the energy they use. It owns the high voltage (transmission) electricity network in England and Wales and operates the system across Great Britain – it does not cover Northern Ireland. National Grid also owns and operates the high pressure gas (transmission) network as well as owning four of the eight gas distribution networks.

On a cold winter’s day the electricity network will deliver around 1.1 TWh*, while the gas network can deliver five times that amount of energy. The two networks have quite different characteristics, with electricity able to travel the length of Britain almost instantaneously while gas takes many hours, leading to an analogy being drawn with the comparative inertia of a motorbike and a super-tanker. For this reason, the two networks are discussed separately.

In the drive to cut carbon dioxide emissions, Britain will invest £110 billion in decarbonising the electricity system between now and 2020. These ‘greener’ sources of power will include wind turbines, solar photovoltaic and, potentially, nuclear power plants. While low carbon generation will reduce the nation’s carbon emission count, National Grid will need to maintain the reliable and secure power supplies that customers have today. Given the dramatic scale of change scheduled for 2020, the UK government has set aside over £30 billion to upgrade and extend the existing transmission networks.

Rebuilding the electricity network

The electricity transmission system was built to allow spare capacity to be shared rather than replicated within each local network. This permitted the running of the cheapest generation to be maximised and meant larger, more efficient power stations could be built further away from electricity demand. A transmission system requires an operator to manage – or ‘balance’ – the system. Balancing involves matching supply and demand on a real-time basis, ensuring the frequency, power flows and voltages stay within operational limits. As the network cannot store energy, any imbalances must be corrected immediately.

The government has ambitious renewable energy targets. It wants 15% of all energy (around 30% of electricity) to come from renewable sources by 2020, up from just a few percent today. And there is an additional target to cut CO2 emissions by 80% before 2050. Up to 2020, the lion’s share of the renewable contribution will come from wind energy, with current estimates predicting that 30,000 MW of wind generation will be installed by the end of this decade. At the time of writing, around 6,000 MW has been installed.

This significant increase requires the installation of thousands of wind turbines by generating companies either at remote onshore locations or out at sea, and connecting them to the existing network. While hundreds of offshore turbines have already been erected, these have been placed near-shore or within a few miles of the shore. Now, arrays of turbines, or wind farms, will be sited more than 20 miles offshore and in waters tens of metres deep.

For the first wind farms, a single subsea cable was installed to link back to the onshore grid. Decisions are now being made as to whether a more cost-effective approach would be to construct a few high voltage, offshore cables to bring ashore the power from several wind farms. National Grid’s estimates indicate that this approach will prove to be both cheaper and more reliable in the long term. Traditional high voltage direct current (HVDC) links require a strong AC system at either end and deliver power by injecting current into the receiving system in synchronism with the local voltage. Voltage Source Convertors (VSC) are now being introduced that give the ability to control the voltage at the receiving end. The advantages are that the power flows on multi-terminal links can be controlled and there is no need for both ends to be live before a link can be brought into service.

In addition to laying these new undersea cables, the electricity network on land will need to be strengthened and extended to accommodate the rising influx of renewable power generation. In particular, there is a shortage of transmission capacity between Scotland and England owing to the investment in renewable energy north of the border. National Grid and Scottish Power will provide some of this new capacity by building a subsea link from central Scotland to north Wales and there are plans for a second link on the east coast if required. The cost of HVDC convertor stations at each end make the technology more competitive for longer routes, and subsea cables also reduce the planning issues associated with new overhead lines. HVDC also has the advantage that the flow on the route can be controlled independently.

The three transmission companies are National Grid, Scottish Power and Scottish Hydro, with National Grid being the system operator for all three. These companies are responsible for extending the existing energy network to accommodate new nuclear plants and gas-fired power stations. Generation is the production of electricity from another source of energy (coal/gas etc) and transmission is the bulk transfer of the electricity around the country. Distribution networks then deliver the energy to end users.

It is the generating companies who make forecasts of costs and revenues to decide what type of new power plant will be built, when and where. When the companies plan a new station, they approach National Grid for a connection. However, nothing is certain until contracts are signed. For longer-term planning, National Grid has developed three scenarios to identify plausible outcomes. These scenarios currently predict that between 7 GW and 20 GW of new gas plant will be built by the mid-2020s, and we expect new nuclear to start commissioning in the early 2020s.

The scale of works is huge. National Grid and the Scottish Transmission Companies have identified the potential scale-up needed to accommodate not only a wide range of new generation which will meet the UK’s renewable energy targets but also other essential new generation that will be required to maintain the demand/supply balance. The total estimated cost of the potential reinforcements is around £8.8 billion. The current transmission network has almost 80GW of plant connected, but after the scale-up the resulting network would be able to accommodate a further 38.5 GW of new generation, of which 24 GW could be a combination of onshore and offshore wind generation.

As well as building and replacing electricity infrastructure, the second challenge is to keep control.

Regulating energy supply

Historically, fossil fuel power stations have generated the majority of the nation’s power. The output from these plants is predictable, and importantly, ‘dispatchable’, meaning that these power stations can be instructed by the grid to start up, shut down, or vary output as required. This output was supplemented by a number of nuclear generators designed to operate continuously at maximum output and which were not required to vary their output to meet demand.

These traditional fossil fuel and nuclear stations use ’synchronous generators’ whereby the system frequency and the speed of rotation of every generator are the same. This synchronous generation has its own inertia, because of the large rotating turbines that generate the power, so when there is a mismatch between supply and demand, this inertia stabilises the system frequency.

However, this needs to change. Firstly, with wind and solar generation dependent on the weather, their output is both intermittent and uncertain. National Grid has to take a view of both the expected output and the lowest credible output to ensure that sufficient alternative sources of energy – reserves – are available. As a fossil generator generally takes three to four hours to start up, a decision needs to be made at least four hours ahead of real time.

Secondly, neither wind nor solar use synchronous generators. In both cases, the energy is delivered to the grid via power electronics which provide a continuous output and do not contribute to stabilising the system frequency by delivering more energy as the frequency falls, or absorbing it as the frequency rises.

The implications of using more intermittent generation on the energy network pose a variety of challenges. Despite increasing uncertainties in power generation, the entire electricity grid needs to give customers a stable supply of 50Hz at the appropriate voltage. Traditionally, a grid operator has been able to regard generator availability to be reasonably settled, with system demand being the main uncertainty. Looking ahead, the output from renewable generation will be the biggest unknown, which in turn will drive uncertainty into the quantity of fossil generation required and the flows on the transmission network.

Tackling wind imbalances

Graph of daily load curve illustrating the uncertainty associated with intermittent renewable generation

Graph of daily load curve illustrating the uncertainty associated with intermittent renewable generation

The expected increase in wind and solar generation will have the greatest impact on the way operators balance the energy network. Even a few hours into the future, the level of output from wind turbines is uncertain. National Grid anticipates that by 2020, 70% of the forecast level of wind can be relied on, which requires a contingency plan to cover the remaining, uncertain 30%.

Flexibility has always been needed to deal with generation losses or errors in demand forecasting. For example, pumped-storage hydroelectricity power plants are used for load balancing. The method stores energy at times of low demand in the form of potential energy of water, pumped from a lower elevation reservoir to a higher elevation. During periods of high electrical demand, the stored water is released through turbines to produce electric power. Operating gas or coal plants at less than full capacity can also introduce more flexibility into the system. This is known as part-loading and can be adjusted depending on how the wind changes.

However, the use of part-loaded plant both lowers the efficiency of the plant affected and requires sufficient demand on the system to accommodate all the part-loaded plant without supply exceeding demand. This latter point can be a particular problem at night when demand is lower.

We at National Grid have considered four potential ways to deal with this. First, we have invested in improved wind forecasting, to have greater confidence in the level of wind generation that will occur. Second, we have looked at using purpose-built generation that can deliver output within 10-20 minutes from a standing start, such as flexible gas turbines. Third, if more storage connects to the system, it could be used to store any surplus energy, although many storage technologies are at the development stage and are not yet suitable for large scale deployment. Lastly, we could export surplus energy to neighbouring countries via undersea electricity cables – interconnectors – that link one nation’s electricity network to another.

Better wind forecasting, while valuable, can never fully eradicate the uncertainties over wind predictions and hence cannot be a complete solution. Analysis suggests that additional gas turbine stations or storage are currently too expensive given that the equipment will only be used occasionally. However, the fourth option, interconnectors, is looking more promising. If, for example, there is more wind than predicted, the amount of power being exported over the interconnector to a neighbouring country could be increased (or conversely the level of import reduced). But, if the wind drops, the interconnector could return to its previous transfer, in effect, providing a backup power supply. The interconnector would also reduce the need for part-loaded coal- or gas-fired power stations and can earn revenue arbitraging between the two markets as well as providing flexibility. In effect, the scale of intermittency is driving a move from performing a national optimisation of generation, towards a working at a more European level.

Interconnection is not a new concept although previously it has generally been used for planned transfers rather than to create real-time flexibility. Several interconnectors already exist between Britain and other nations including France, Eire and the Netherlands, and more are planned. National Grid is in talks with Norway about such a link and has invited third parties to put forward other proposals.

However, more ambitious plans are possible. Why not link the many offshore European wind farms with transmission cables, and form a Europe-wide offshore electricity network? The UK is currently working with countries including France, Germany, Norway and Sweden to negotiate the North Seas Countries Offshore Grid Initiative. This planned network of underwater transmission cables would link offshore wind farms to other power sources in nearby countries.

In the future, surplus wind energy produced off Britain’s coast could be exported to Norway and be used to pump water in the nation’s hydroelectric power stations. Meanwhile, the electricity produced by hydropower could by then be sent to the UK when the demand was high, but the wind was low. The project, still very much in its infancy, could pave the way to a broader European grid stretching from Ireland to the Baltic states and as far south as North Africa. The concept clearly offers possibilities, but faces questions of ownership, as well as many technical hurdles, such as ensuring that a widespread failure on the continent could not lead to a shutdown of the GB system.

National Grid is creating 20 miles of cable tunnels under London to avoid the disruption associated with traditional cable laying techniques © National Grid

National Grid is creating 20 miles of cable tunnels under London to avoid the disruption associated with traditional cable laying techniques © National Grid

Maintaining network security

As well as balancing the system, National Grid has to ensure that its network is secure at all times against all credible system faults – broadly the loss of any single piece of equipment, or the complete loss of an overhead transmission route. Following such a fault, any overload must be managed within the short-term capacity rating, system voltages must remain within tolerance and any oscillation in power flows must be dampened rapidly. Historically, if the network was secure at times of peak demand, then it was generally secure at lower demand levels. But given the increase in wind generation, maintaining network security will be more difficult as operators now have to consider a range of potential power flow patterns.

National Grid has always had to manage the impact of faults and, as system transfers have increased, we have strengthened the system by uprating equipment, installing automatic voltage control equipment and quad boosters. A quad booster is a transformer that can increase or reduce the flow through the overhead line or cable to which it is connected. When there are alternative parallel paths that the electricity can take: a quad booster can be used to manage the flows so as to better match the various flows with the capability or rating of the different paths.

National Grid is working to respond much more quickly to changing power flows on the system and hence the consequent changing challenges to network security. One approach is to enhance the online security assessment software so that, in addition to monitoring the impact of potential faults, it also provides advice on optimising the system. It does this by providing advice on the best settings for quad boosters/voltage control equipment and the appropriate post-fault actions required to relieve any overload.

Using fast-acting automatic post-fault actions is a further option. As described above, a fault on one piece of equipment can lead to an overload on a parallel path. To remedy this scenario, the operator can manually reduce the power being delivered from a power station, although this takes around 10 minutes, during which time the network must operate, while tolerating the overload.

If an inter-trip scheme is installed, it can continuously monitor the transmission system and automatically trip the generation should the relevant fault occur. By reducing the overload almost immediately, the network is able to tolerate higher pre-fault power flows, and more power can be transferred.

Network operators are likely to use each of these potential responses to help manage system constraints. In fact, National Grid will have to use all the ls at its command to respond to the changes that are coming, and while some of those tools are familiar, the challenge is to respond to the rapidly changing network flows.

Managing gas

Just as the electricity system requires the frequency, voltages and power flows to be managed, the gas network needs to be balanced to maintain pressures within acceptable limits. Today’s gas transmission network was originally designed to operate with a constant flow, matching production from the North Sea. Nearly all the gas used to come from production on the UK continental shelf, bringing stable, predictable flow rates. This is no longer the case. The nation now imports significant volumes of gas from Norway, the Netherlands and Belgium via pipelines (also known as interconnectors). Gas is also imported as liquefied natural gas, via terminals at Milford Haven and the Isle of Grain. Storage sites are also playing an increasing role in storing surplus gas and releasing it at times of shortage. Clearly, as more industry players put gas into the network as well as withdraw it, gas flows into and within the system will become more and more variable.

Short-term imbalances can be met by allowing the pressure in the system to vary, although gas only travels through the transmission system at around 20 mph, so imbalances in the gas pressures can take many hours to resolve. Historically, the cumulative mismatch between supply and demand has occasionally reached 300GWh. As gas flows within the network become more and more dynamic, we anticipate that, by 2020, these mismatches could double or even treble.

The increase in wind power on the electricity network will also result in more dynamic gas flows within the gas network. Managing the electricity grid to accommodate intermittent generation will involve varying the power generated from gas-fired power stations, which will be a major driver of intermittent demand on the gas system. In response, National Grid is studying varying gas flows across the day and is identifying the most credible worst-case scenarios for supply and demand mismatches – for example, renewable generation failing to meet its predicted levels. These scenarios will then be used to assess the likely effects on the gas network. The impact will also depend upon which power station is used to meet the shortfall – the pressure drop is greater if the plant used is on a remote part of the network.

While the electricity network needs investment in new or additional infrastructure to keep power flowing, for gas, the solution is expected to be a mixture of assets and new commercial arrangements. National Grid is in the process of identifying how much flexibility is inherent in today’s gas transmission network, and this involves looking at exactly how today’s network users operate, and how this might change.

The current balancing rules only require participants to balance on a daily basis, reflecting the expectation of constant flows over the day. Going forward, it may be necessary to either tighten the rules to require all parties to balance over a shorter period or for National Grid to develop balancing tools that would require delivery within a shorter time scale. Either way, each option would provide the operator with tighter control over the network.

Another consideration is that, in the past, when gas was in short supply, generating companies could switch to using more coal- or oil-fired stations, releasing gas supplies, especially during the winter months when demand for gas is high. In effect, coal stocks at power stations offered significant support to the security of the gas system. However, environmental legislation has now set absolute limits on emissions of sulphur dioxide and nitrogen oxides from power stations, so we will see many of the coal- and oil-fired power stations closing in the coming decade. This will increase the dependence of the UK as a whole on imported gas.

So whether you look at electricity or gas, it is certainly a time of change for the industry. At National Grid, we join everything up and that puts us at the heart of one of the greatest challenges facing our society: the creation of new sustainable energy solutions for the future and the development of an energy system that can support our economic prosperity in the 21st century.

[Top of the page]