Coping with wind

This article was published in Ecologist on 27 April 2010, and in the June edition of Energy World.

Texas is full of surprises. In the historic home of the oil industry, the electricity supply is going green. A landscape that for over a century has been carpeted with ‘nodding donkey’ oil wells is now sprouting wind turbines. The state has 9 gigawatts (GW) of wind generating capacity, more than double that of Britain, providing 6% of its electricity last year. If Texas were a country its wind capacity would rank fifth in the world, behind only Germany, China, Spain and India. Grid operator ERCOT is building power lines to connect another 18GW, which would raise wind capacity in the Lone Star State to over 40% of peak demand.

The Texas wind-rush is surprising not only because it confounds the stereotype, but also because the rapid growth of wind capacity means that quite often wholesale power prices in Texas turn negative. In other words, generators are forced to pay their customers, the electricity supply companies, to take their power. This tends to happen when the wind is blowing strongly during off-peak hours and there simply isn’t enough demand for all the electricity produced. The problem is exacerbated because coal and nuclear baseload generators cannot easily ramp their output up and down, and because West Texas – where most of the wind turbines live – is only poorly connected to neighbouring grids, so cannot export much of the excess power. The frequency of negative pricing has soared in line with growing wind capacity, from under 100 times in 2006 to over 3000 in 2009.

For customers it may sound like a nice problem to have, but if this trend were allowed to worsen, it would eventually undermine the business case for building more wind farms. It’s also an early-warning light of the fundamental problems of a growing reliance on wind power: that the wind sometimes blows when we don’t need electricity; and sometimes doesn’t when we do; and even when it’s blowing a gale and we want the power, wind farm output is variable and has to be balanced in some way to keep the grid stable. All this presents both financial and technical challenges that are bound to become more acute as the proportion of wind generation rises – as Britain may be about to find out.

Earlier this year the government awarded licences to build 32GW of offshore wind capacity, enough to provide a quarter of our yearly electricity, more than any other country has yet achieved, by 2020. So Britain will have to solve the problems of integrating an unprecedented proportion of wind into the grid within a decade. The challenge is to cut emissions from electricity generation while keeping the lights burning without breaking the bank. Some analysts claim it can’t be done, and the very attempt will threaten our security of supply. But luckily several European countries are way ahead of us, and pioneering a range of approaches could eventually lead to a totally renewable electricity supply.

One of the most ambitious is Spain, where wind capacity has soared in recent years under a system of generous feed-in tariffs. Capacity stands at 19GW today, generating around 14% of the country’s electricity in 2009, and once – on a windy Sunday night last November – briefly delivering 54% of its power. Yet the government wants to go much further and has set a target of 29GW by 2016.

For a country so determined to grow its wind capacity, it is ironic that Spain has discovered one of the most important factors is the ability to shut wind farms down, or at least reduce their output from time to time. To achieve this, the Spanish grid operator REE (Red Eléctrica de España) has built the world’s first renewable generation control centre, the Centro de Control para el Régimen Especial, or CECRE, housed in an anonymous concrete campus in the shadow of Madrid’s Barajas airport. According to Miguel de la Torre, the REE official who shows me round, CECRE has been crucial to Spain’s success in incorporating so much wind power so far and its future plans.

Controlling the power output of wind farms is important for keeping the system within its safety margins when the wind is blowing strongly and demand is low, and all the more so as wind capacity grows. Yet in Britain, the National Grid control room at Wokingham cannot even measure the output of half the country’s wind farms, still less control them. By contrast technicians at CECRE receive live output data from every wind farm once every 12 seconds, displayed on a wall of huge screens and maps, which helps their colleagues in the main grid control room balance the variable wind output by raising and lowering the production from flexible generators such as gas fired plants.

Wind turbines have priority in the Spanish system, but sometimes output has to be curtailed, and CECRE can send signals back to the wind farms requiring them to trim production within 15 minutes if necessary. That’s important because if they couldn’t control wind output so quickly, REE would have to set production limits a day or more ahead on the basis of less reliable weather forecasts, and allow for a larger margin of error. The control offered by CECRE means they can run the system nearer to its limits, using more wind power overall. “Normally we can allow more wind onto the system because we know we can reduce the wind output quickly if we need to”, says de la Torre.

This and other technical reforms will allow Spain to raise its wind capacity by another 9GW (or one Texas) in the next six years, while maintaining a stable grid, says Javier Revuelta of REE’s planning department. “We are very confident the system will work” he says, “We know that some days we will have to curtail the wind output, but it’s not a technical problem for the system”. This much capacity could supply up to 70% of Spain’s instantaneous power on occasion, estimates Revuelta.

The ability to turn turbines down is all very well when there’s more wind power than demand, but what about when the wind drops? Spain’s answer is to ramp up production from fast-reacting gas fired power stations, and this could easily end up being the default position for Britain. But Portugal has come up with a more climate-friendly alternative: a massive expansion of hydro electricity. Hydro power is the perfect partner for wind, because reservoirs and dams have one thing turbines lack – the ability to store energy.

Portugal already has a lot of wind and hydro, but most of its hydro is ‘conventional’, meaning the reservoirs are rain-fed and the water can only flow downhill once. To make the most of the wind-hydro partnership, the hydro should be ‘pumped storage’, where two reservoirs at different heights are connected by pipes and reversible water turbines. Then when the wind blows at night and demand is low, cheap electricity can be used to pump water uphill, and during the day when demand and prices are high, the water can rush back down to generate power. It takes more energy to pump the water up than is generated on the way down, but that’s the price of storing energy so it can be used when needed, rather than when the wind chooses to blow.

Portugal plans to more than double its wind capacity from 3.5GW to 8GW by 2020. To help balance this it will also increase hydro from 5GW today, of which less than a fifth is pumped storage, to 9.5GW by 2020, when around half will be pumped storage. This will lift the renewable proportion of Portugal’s electricity from 45% today to 60% in 2020, despite a big predicted rise in demand. “We are very lucky”, admits Josée Medeiros Pinto, deputy director of planning for REN, Portugal’s grid operator, “if we didn’t have hydro, hitting our renewables targets would be much more difficult and expensive”.

Britain is not quite so lucky. True, we do already have 1.4GW of conventional hydro and almost 3GW of pumped storage, at plants such as Dinorwig in Snowdonia, mostly built decades ago to mop up excess nuclear power during off-peak hours. But the potential to increase capacity is limited. Scottish & Southern recently announced plans to build another 900MW of pumped storage at Loch Lochy and Loch Ness, and it’s estimated Britain could build another 2GW of conventional hydro. But that’s not remotely enough to balance 32GW of wind.

Another way to balance wind is to trade it with your neighbours, which is how Denmark has coped with a huge increase in wind capacity in recent years – although this strategy is starting to creak. Because as well as lots of turbines Denmark also has thousands of combined heat and power (CHP) plants, which are efficient but inflexible. The plants provide district heating and electricity from the same equipment, so if you need the heat you get electricity too, but if the wind is blowing and demand is low, the result is often too much power. The solution so far has been to export excess electricity through interconnectors to Germany and Norway when the wind blows, and import when it doesn’t.

Source: Energinet, Incoteco

On a graph, Danish wind production and electricity exports correlate very closely (see graph) and critics of Denmark’s energy policy such as energy consultant Hugh Sharman argue this makes a mockery of claims that the country generates 20% of its electricity from wind. Defenders of the policy say it is the inflexibility of coal and CHP that is to blame, and it is this power that is being exported when the wind blows.

Whoever is right – if it even matters – it seems certain Denmark will not be able to rely solely on international trade to balance its grid as it pursues ambitious plans to double its wind capacity by 2025 and generate half its annual electricity from wind. That’s because a huge increase in wind farms is also planned in Germany and Norway, so when the wind blows in future, the neighbours won’t be able to absorb the excess power from Denmark. The Danes know they have to find ways to balance the grid within their own borders.

The problem is not trivial. Already wind generation occasionally exceeds total demand in off-peak hours, and with twice the capacity it would often exceed even peak demand (see graph). Network operator Energinet has launched an industry-wide project to develop an ‘EcoGrid’, where wind power is balanced not only by other forms of electricity supply but also through demand management – shifting consumption to suit supply, rather than the other way around. This is a “major challenge” that demands “profound changes in philosophy” according to Kjeld Norregaard of the Danish Technology Institute (DTI), which leads the research.

West Denmark currently generates the equivalent 25% of its yearly electricity consumption from wind, and already wind power occasionally exceeds demand during off-peak hours (left hand graph). If capacity and generation double by 2025 as planned, wind generation would exceed even peak demand much more often (right hand graph). Source: Danish Technological Institute

In an initial report, EcoGrid researchers found that while Denmark’s CHP is part of the problem, it may also offer part of the solution. CHP plants come equipped with hot water storage tanks, and these could be fitted with electrical heaters to soak up excess wind power and save it to be used later as heat. The next step would be to install hundreds of thousands of ground source heat pumps in houses and buildings across the country, which would perform the same role but with greater capacity. Another idea is a shift to electric vehicles, which would have some flexibility to charge when the wind is blowing and electricity cheapest. Already a €103m joint venture between the Israeli company Better Place and Danish energy giant DONG is working to roll out tens of thousands of electric cars and a recharging infrastructure.

These measures would be just a start, and EcoGrid estimates that total flexible demand capacity could amount to 1.3GW, or about a fifth of Denmark’s peak demand. To begin with, demand management would depend on big commercial electricity consumers such as supermarket chains, but eventually could include every household as part of a ‘smart grid’.

Domestic demand management measures are also being developed in Germany, where, like Texas and Denmark, electricity prices regularly go negative, and where energy market reforms are spurring change. If the pilot project I visited in Mannheim, near Frankfurt, is anything to go by, this approach could not only to raise the proportion of renewables on the grid, but also cut customers’ bills.

In a neat suburban street of well-insulated, steep-gabled houses, one sign of the gathering energy revolution is that Harry Wirth, rather than his wife Antje, now hangs out the washing. Herr Wirth works in the planning department of the local energy supplier MVV, which is also being forced to change its ways. Traditionally Germany’s 800 electricity companies have sold power at a fixed price per unit, but from next year they must offer their customers flexible tariffs that reflect the changing wholesale price of energy throughout the day. That prompted MVV to experiment with demand management, and Herr Wirth to volunteer his family as guinea pigs, with interesting and unintended consequences around the house.

The MVV trial is based on a gizmo called the Energy Butler. About the size of a domestic wireless router, the Energy Butler wirelessly controls four smart mains sockets that supply the washing machine, drier, dishwasher and fridge-freezer, and connects to a secure website via the internet. On the website, MVV publishes a forecast of hourly electricity prices for the following day, which the family used to decide broadly when each socket should be supplied with power. Since prices are typically lowest overnight – when demand is low – they could set the appliances before going to bed, knowing the Energy Butler would chose the cheapest time for them to run while completing the cycle by morning.

After a few days’ use, the Energy Butler develops enough intelligence to ‘set and forget’, and future versions should be smarter still. Meanwhile, the fridge-freezer has an additional thermostat installed to ensure the maximum possible cooling is done during times of cheap power. And the system can always be over-ridden at the flick of a switch.

The Wirths found the Energy Butler changed their energy use in several ways. The dishwasher was used overnight rather than during peak hours. The website’s graphical display of the energy consumed by their clothes dryer shocked them into giving it up altogether. And instead of all doing their washes on ‘wash day’, the family of three changed to one per night several times a week. Antje didn’t like the wet clothes sitting in the machine for hours in the morning but had no intention of getting up at 5.30 to hang them out, which is how Harry got the job. “That wasn’t part of the plan”, he says, laughing.

The family’s electricity bill fell from €900 to €810, and would have fallen further had they tried harder, says Harry. For the company, it is early days in a pilot to test customer reaction, expanding to 1500 homes next year, but with a clear end in sight. “With a fully workable solution, the result would be far more renewable energy on the grid without it becoming unstable”, says MVV spokesman Dirk Pohlman. The company sees lower emissions and prices as a great advantage in an increasingly competitive market.

The Energy Butler system is completed by another smart little box of tricks that sends automatic readings from the household’s electricity meter through the mains cables all the way to MVV headquarters. This box is supplied by a local company PPC (Power Plus Communications), which has installed similar equipment at electricity substations throughout Mannheim, giving MVV a detailed picture of power flows at a street by street level for the first time. This is important because variable generation comes not only from remote wind farms via the national grid, but also from myriad solar panels on local roofs, and this creates its own problems.

Some German neighbourhoods now have so many panels, explains PPC chief executive Eugen Mayer, that on a sunny day, instead of drawing energy down from the higher voltage regional grid, they pump power back up into it. Since the system equipment wasn’t designed to work this way, it’s important to balance supply and demand at a local as well as national level, and that makes the Energy Butler and PPC communications systems even more vital. “It’s impossible to double renewable supply in Germany without demand management”, says Mayer. But who knows, if demand management allows more solar panels as well as wind turbines, Frau Wirth may be able to go back to washing during daylight hours.

In Britain, where power demand ranges between about 20GW and 60GW depending on season and time of day, the National Grid estimates the potential flexible demand could amount to 12GW by 2020 (see graph). But that assumes a fleet of 1 million electric cars, and the company thinks under 8GW is more realistic. Much will depend on smartening the grid, yet the government’s deadline for the installation of smart meters in every home – a basic building block – is ten years away.

There could be as much 12GW of flexible demand to help balance the grid by 2020, but National Grid thinks under 8GW is more likely. Source: National Grid

While demand management is clearly vital, it is not much help for the biggest problem of all: when the wind fails to blow for days or even weeks, as during the big freeze in January. Even countries that are way ahead in wind balancing like Spain and Portugal continue to rely on fossil generation to fill this gap. In Denmark, Kjeld Norregaard freely admits that so far “we have only solved a fraction of the problem”.

So how will Britain crack it? Conventional thinking suggests in the short term we will follow the pack. In one planning scenario, ‘Gone Green’, which National Grid describes as ‘plausible but extremely challenging’, 28GW of wind in 2020 is combined with 12GW of new gas fired power and 3 each of nuclear and coal – perhaps with carbon capture. But an alternative, entirely renewable solution is beginning to take shape far more quickly than anyone could have anticipated.

Two years ago the idea of a continent-wide supergrid seemed science fictional, but today elements are suddenly falling into place. Last December eleven North Sea countries signed a memorandum of understanding to establish a superefficient high voltage direct current (HVDC) sub-sea grid, largely to trade the output of planned offshore windfarms. Officials from energy ministries, regulators, utilities and the European Commission met in Brussels in February and March and are preparing a detailed action plan to be signed off by the end of the year. In a separate development, ten major power engineering companies – including giants such as Siemens and Areva – launched the Friends of the Supergrid (FOSG), to demonstrate industry confidence that this massive infrastructure project is achievable.

The supergrid would solve the ‘Danish problem’ of reliance on trading power with your immediate neighbours because the distances involved are so much greater and encompass different weather systems. With electricity trunk routes stretching from Ireland to Kazakstan, and Scandinavia to Morocco, the wind would always be blowing somewhere, and there would always be demand for that power somewhere else. “I look forward to the day when Scotland turns on the kettle to be powered by North African electricity”, said John Sturman of Parsons Brinckerhoff, the engineering consultancy at the FOSG launch in London in March. With so many different wind farms – and solar and hydro plants – feeding into the grid, the variations in output would even out to create a far more dependable supply. Renewables would be balanced not by fossil plants, but by each other.

Matthew Knight of Siemens explained there were no major technical hurdles left to overcome, but said it was critical that industry standards be established to prevent a VHS-Betamax-style battle delaying progress. Other major questions must still be resolved, however, including who will own, finance, regulate and operate the new grid. But FOSG expects governments to establish a North Sea Transmission System Operator by 2013, and the first cables to be laid by the time Britain has built its 32GW of offshore wind. Indeed, Eddie O’Connor, chief executive of Mainstream Renewables, the driving force behind FOSG, says Britain cannot build that much offshore wind capacity “reasonably and cheaply” without it. It may be a stretch, but that means the first elements of a truly sustainable way to balance huge amounts of renewable generation across Europe could be operating by 2020 – with a following wind.