Impact of renewable energy on UK power transmission costs

From Wikipedia, the free encyclopedia

It is often claimed that a large increase in wind power or other renewables would cause a massive rise in the cost of transmission in the United Kingdom.

As of 2006, the transmission costs of the entire UK high voltage National Grid are presently in the region of 0.2p/kWh. If it were the case that the entire present grid had to be duplicated to accommodate close to 100% wind, then this would add an extra 0.2p/kWh to the cost of power. It may be that the grid would have to be tripled, or even quadrupled, but this latter would on this basis, only add 0.8p/kWh.

The cost of power for offshore wind farms usually explicitly quote the cost of connecting the farm into the National Grid.

Contents 1 The cost of backup power for intermittent sources 1.1 Back up costs as represented by Spark and Dark spreads 1.2 Values of Spark Spreads and Dark Spreads 1.3 Wind generation 1.4 Transmission network and grid reinforcement costs 1.5 Total increased costs of back-up, transmission and balancing 2 See also

[edit] The cost of backup power for intermittent sources

Much speculation is published about the high cost of backup power for intermittent sources such as wind power when connected to the national grid, but it is easy to demonstrate these costs are not high, relative to the delivered cost of power, using published information, which is thus not open to dispute or speculation.

The point is that the backup power stations for even 100% windpower penetration have already been built – it is the CCGT, coal fired and nuclear power stations that already exist, so there is no capital cost to providing this mythical backup.

The cost of keeping this plant idle is also very low as we shall see (except for nuclear). To understand this, firstly it should be borne in mind that 96% of the cost of running today’s conventional fossil fuelled power stations is the fuel cost. So if they are not operated, 96% of the costs disappear. If overnight, sufficient wind turbines were constructed to target close to 100% of the electrical energy used in the UK on an annual basis, then the cost of paying the owners to keep the power stations in reserve to provide backup, is simply the money or profit power station owners would have expected to make over the coming year and the rest of its life from buying fuel and selling the resultant electricity. If they are paid this profit, then they are financially no worse of than if they had run. During those periods when there is insufficient wind to meet the demand and the power stations are required to make up the difference, then we will be back to the pre-existing situation and costs will revert to what they would have otherwise have been – ie the fuel costs.

[edit] Back up costs as represented by Spark and Dark spreads

This profit is readily available in published specialist energy industry news services such as Spectron, Bloomberg, Reuters, Argus, Platts etc and is referred to as the Spark Spread for gas fired power stations, and Dark Spread for coal power stations. These published spreads are the difference between what the power station pays for its fuel, and what it sells the resulting electricity for. In other words it is the profit the owners expect to obtain which they will use to pay off the capital and maintenance costs on the power station and earn their return. Zero spark spread means the power station is merely breaking even, and negative means it is losing money. When Dark Spreads fell to zero a few years ago, this led to the bankruptcy of the owners of the 4 GW Drax power station, which then ended up in the hands of the banks.

[edit] Values of Spark Spreads and Dark Spreads

It is generally agreed amongst power industry experts, that spreads need to be about £12/MWh for new gas fired stations, and £14/MWh for new coal fired stations. Presently levels are around £7.5/MWh or 0.75p/kWh and no one is apparently complaining too much.

An individual power station load factor might be on average 43% (that is they are only working for about 43% of the time, so they have to earn their profit – their spark or dark spread within that time). Thus each year the owners of a large CCGT with a 43% load factor are happy to earn profit equal to £7.5 x 8760 x 0.43 = £28,250 per 1 MW.)

If we imagine a scenario in which we aim to generate 100% of power used in the UK, say 3.6 ×1011 kWh, from wind generators then we would need to construct about 102GW of wind plant, assuming an offshore annual load factor of 0.4 for the turbines. (Due to windless periods we might only manage 95%, without special measures see later)

To keep the entire fleet of present power stations available for say half an hour when there is no wind anywhere in the UK, they would need to be paid the spark spread they would have otherwise earned before the turbines were built. Since our wind farms our now exactly generating the kWh the conventional stations would have been generating then clearly the wind farms would have to pay in some way, the same spark spread to the power station owners for each kWh they generate, so the power station owners would be no worse of than before..

In other words the cost of providing back up for a 100% wind target is in fact the spark spread or a mere 0.75p/kWh – about 1/10 the present cost of power to consumers..

[edit] Wind generation

Much less than the present power generation capacity would be needed for close to 100% wind generated target.

In fact the cost of these Spark Spread payments would be significantly lower than the present power generating fleet because much less of the present generating capacity would be required for the following reasons:

• As Graham Sinden of Oxford University’s work has shown, there was not a single hour over the last 30 years when there was no wind simultaneously over the whole of the UK. So 102 GW of wind would always be producing some power – several GWs – in other words we would not in fact need to retain anything like the present 70GW entire fleet of power stations.

• As pointed out earlier there is already 20 GW of unused capacity for this purpose, comprising standby diesels which are already paid for and therefore need only a fraction of this spark spread, to operate, and could be brought in to some variant of Reserve Service.

• EHVDC (Extra High Voltage Direct Current) power cables are now relatively cheap and have very low losses. This means we can readily export power to Europe during windy periods when the 102 GW capacity may then exceed the normal 60 GW peak demand on the national grid (national grid peak is of course usually less than the 60 GW peak), and re-import during low wind periods. 102 GW of wind capacity will rarely be producing this output since it is rarely very windy everywhere simultaneously - again see Sinden's work.

• EHVDC losses are only 3% per 1000km, the longest one currently is 1500 km, and a 500km sub sea cable is presently being constructed to link Norway to Holland. Central Germany has been connected to Norway’s hydro stations since the 70s, and Norway’s existing 25GW hydro could easily be converted to pumped storage.

• There is no reason why something similar to Frequency Response could not be applied to a 1 kW deferrable load or load brought forward in each of the 20 million UK homes. Equipment such a freezers, fridges, storage heaters, immersion heaters would give 20GW instantaneous load shedding equivalent to Frequency Rsponse(See David Hirst of Responsive Load for further information on this point)

• 1 kW Stirling micro-CHP are currently being introduced into domestic properties. These could be fitted in each of the UK’s 20 million homes and would give 20 GW of callable generation capacity equivalent to Reserve Service

• An all electric car has 70 kW motor and associated batteries. Assume there are 1 million within 20 years, (present car population is about 20 million) which are only in use for about 2 hours per day. For the rest of the time they will be on charge either at home or connected to public charging points which are springing up even now in streets and car parks. Because these technologies would be pre-existing for other purposes these would need a lot less than present spark spreads to justify their involvement.

These car batteries collectively will act as a gigantic storage battery with instantaneous output or input absorption capacity of 70 GW – perfect for storing peaks in wind power. Most commentators do not realise that the installed capacity of vehicle engines in the UK is an astonishing 1800 GW which is 25 times the capacity of the national grid. Because these would be pre-existing for other purposes they would need a lot less than present spark spreads to justify their involvement.

• Assume in 20 years time there are, five million hybrid electric vehicles, each with a 20 kW motor and auxiliary engine generator. Since these will also be on charge whenever parked, then the engine + generator can also be used as part of some form of reserve service, as cover for extreme windless periods. Five Million 30 kW hybrids would have an instantaneous power output from fossil fuels of 199 GW. This would only need to be called upon in those very rare days when there is no wind at all anywhere in the UK. Again – because these would be pre-existing for other purposes these would need a lot less than present spark spreads to justify their involvement. They will also of course have a massive power absorption capacity during windy periods.

• William Orchard has proposed the installation of a heat network to achieve diversity of heating and domestic hot water and the installation of 500kW gas fired CHP generators with a standby fuel connected to the LV side (415V) of the numerous utility 11kV to 415V transformers. These are capable of being actively managed and would be ideal for matching wind power. The ability to readily store heat from surplus windpower gives significant advantages for this option to achieve national carbon savings for existing buildings to meet National carbon dioxide targets.

Such a system would also secure local electricity and heat supplies when terrorist or military action results in loss of the High Voltage transmission system and 11kV supplies.

Thus it can be seen that the 0.75/kWh back up charge is very much an upper limit, even for close to 100% windpower penetration, and can be expected to be in practice much lower.

[edit] Transmission network and grid reinforcement costs

Again people erroneously talk of the massive cost of reinforcing the grid to cope with fluctuating renewables. What is lost sight of is that the cost of the grid, i.e. the transmission costs in the electricity bill is tiny, compared to the fuel costs which dominate, and is again published – it’s explicit in most non-domestic power bills. The various elements are:

Tuos - Transmission Use of System charge – This is in fact what we earlier called the Triad charge – in the South West its about £21000/MW / year, but on average about £15,000 / MW /year.

If averaged over all power supplied in the UK then this is currently around 0.2p/kWh. This is calculated by taking the total annual triad charges, which are say £15/MW/ year x 50 GW = £750m and dividing it by the total number of units sold per year – say 3.6 × 1011 kWh.

Assume you had to duplicate the entire grid to accommodate wind (extremely unlikely) – then this would still only add an extra 0.2p/kWh to the cost of power.

• Duos - Distribution Use of system charges - would not change - about 0.2p/kWh.

• Bsuos - Balancing Services Use of System - currently around 0.2p/kWh - hard to see why that would get much larger – but say it increases by 50%.

[edit] Total increased costs of back-up, transmission and balancing

So we can now add all these increased elements up:

• Spark spreads – 0.75 p/kWh
• Extra TNUoS / TRIAD assume same again - 0.2p/kWh
• Distribution - no increase
• BSuos - assume 50% increase - 0.1p/kWh

This gives a total of 1.05 /kWh – 10 % of present delivered domestic power costs - say 10% of present domestic power costs.

Thus the total extra costs of targeting 100% wind power, (which would in practice only achieve 95%)would probably add about 10% of present consumer costs due to back up and increased transmission, at the maximum.

[edit] See also

http://en.wikipedia.org/wiki/NGT_Standing_Reserve

• Calculating the cost of the UK Transmission network: Cost per kWh of Transmission
• Calculating the cost of back up: See Spark spread