A version of this article was published in New Scientist on 12 March 2009.
Thomas Edison might have relished the irony. Just as his most famous legacy, the incandescent light bulb, heads for extinction, his other great passion, direct current, is set to boom. The bulb that dominated lighting for over a century is now a pariah of climate change and banned in many countries. Meanwhile direct current, which was defeated by alternating current in the race to establish the industry standard during the 1890s, is now emerging as crucial weapon in the fight against global warming, in the form of high voltage direct current (HVDC) transmission lines.
Although the world’s electricity transmission grids are almost wholly AC, it is now becoming clear that HVDC will be crucial to meeting soaring electricity demand and cutting carbon emissions – by transmitting large amounts of power efficiently over long distances and connecting remote offshore wind farms. HVDC even promises to solve the vexed problem of the intermittency of wind turbines and solar panels by allowing the creation of continent-wide ‘Supergrids’, which smooth out the variable generation from many far-flung sources to create a dependable supply. Supporters claim this will make it possible to ditch coal, gas and nuclear altogether and replace them entirely with renewables within a couple of decades.
Elements of a European Supergrid are already beginning to emerge, with plans for offshore HVDC grids being developed in both the Baltic and the North Sea. And political momentum behind the idea is growing: in January the European Commission proposed €300 million to subsidize the development of HVDC links between Ireland, Britain, the Netherlands, Germany, Denmark, and Sweden, as part of a wider €1.2 billion package supporting links to offshore wind farms and cross-border interconnectors throughout Europe. Meanwhile the recently founded Union of the Mediterranean has embraced a Mediterranean Solar Plan to import large amounts of concentrating solar power into Europe from North Africa and the Middle East.
In the US, President Obama’s $150 billion energy plan includes a target of 25 per cent renewable electricity by 2025, implying massive investment in high voltage lines. A recent report from the US Department of Energy found that achieving 20 per cent wind penetration would require new ‘transmission superhighways’ stretching more than 12,000 miles. America’s existing 200,000 mile high voltage transmission network is almost entirely AC, but many of the new lines are likely to be HVDC. According to Dr Graeme Bathurst, technical director of the British grid consultancy TNEI, “Whichever way you look at it, there is absolutely no doubt that HVDC’s time has come”.
It is ironic that Edison lost his ‘battle of the currents’ with Tesla and Westinghouse because in one sense direct current is far superior; it suffers much lower transmission losses than alternating current. That’s because in a DC line the voltage is constant, whereas in an AC line it reverses direction 100 times per second, meaning more energy is lost as waste heat. Because of this, HVDC has long enjoyed a niche role transporting large amounts of power efficiently over long distances. One of the earliest big projects was a 600MW link in New Zealand connecting the north and south islands, built in 1965, and later upgraded to 1200MW.
One disadvantage of HVDC lines is the need for converter stations where they connect to an AC grid. These are big and expensive: for a 3000MW line the converter stations at either end would cover 9 football pitches and cost around $200 million each. But once the link is longer than about 600km, the extra cost is increasingly outweighed by the energy savings, making it economic to transport power over vast distances that with AC would be expensive and technically difficult
The length and capacity of new HVDC projects is rising fast, particularly in China, where lines are being built to transmit hydro power from deep in the country’s interior to consumers on the coast. The Swiss engineering firm ABB has recently been commissioned to build a link from the Xianjiaba dam to industrial Shanghai, which is the world’s longest, at 2000km, and most powerful, at 6.4GW – equivalent to the output of three large power stations. The line’s converter stations will cover 20 football pitches each, but when the project opens in 2011, the company says it will deliver major environmental benefits.
Dr Gunnar Asplund, ABB’s research and development manager for HVDC, explains that enormous amounts of power will be transmitted along a single line of pylons, whereas a traditional AC link would need three abreast. And because the alternative to transporting hydro electricity long distance would have been to build more coal fired power stations near Shanghai, Dr Asplund reckons the carbon dioxide savings could amount to 40 million tonnes per year.
Another major advantage of HVDC is that it can operate over much greater distances underground and under water than AC. That’s because AC produces powerful alternating electric fields that cause large additional energy losses if the line is buried or submerged, whereas for DC this ‘capacitance’ effect is practically negligible. That makes HVDC essential for sub-sea ‘interconnectors’, like the 600km NorNed cable between Norway and the Netherlands that opened last year.
NorNed was intended to reduce electricity prices by sending cheap Norwegian hydro power south during the day, and Dutch off-peak coal and gas fired power north by night. But this year has been exceptionally wet in Norway, making hydro-power so cheap that so far the trade has been almost all one way. That could easily reverse though, says Odd Hoelsaeter, the recently-retired chief executive of the Norwegian grid operator Statnett, “in a dry year like we had in 2002-3, we would be massive importers. Quite apart from the trading benefits, this is a major increase in energy security for both our countries”.
HVDC will also be essential for connecting the more remote offshore wind farms now being planned off Britain and Germany, although this will depend on a new generation known as voltage source converter (VSC) technology, which ABB markets as HVDC Light, and Siemens as HVDC Plus. That’s because, unlike classic HVDC, VSC does not require a strong AC grid at both ends in order to function, and its converter stations are compact enough to fit on an offshore platform.
Perhaps the biggest potential role for HVDC will be to enable the ‘Supergrid’ concept now gaining support in Europe and America. The idea itself is not new – it was first proposed by Buckminster Fuller in the 1950s – but only now is it becoming a practical possibility because of advances in HVDC technology.
Since 2003, Desertec, an organization founded by the Club of Rome and the National Energy Research Center of Jordan, has promoted one version of the Supergrid based largely on concentrating solar power (CSP) in North Africa and the Middle East. CSP is still relatively expensive, but one big advantage is that some of the heat captured during the day can be stored in molten salts and used to generate electricity overnight, and Desertec says this technology alone could supply 17% of Europe’s power by 2050, imported over 20-40 long-distance HVDC lines. But other supporters of the concept argue that the Supergrid could deliver even more – a wholly renewable electricity supply.
The problem with renewable electricity – its detractors claim – is the wind doesn’t always blow, nor the sun always shine. But that is only true of a single location. The wind is always blowing somewhere, and sunrise and sunset come at different times depending on geographical location. So given a large enough grid, the variations in renewable generation should tend to even out, making the supply much more reliable overall.
The potentially huge impact of this ‘spatial smoothing’ has been demonstrated by Dr Gregor Czisch, an energy system consultant, who has made the first quantitative study of how to build an economically viable, wholly renewable electricity supply for Europe and its neighbours – an area stretching from Iceland to Kazakhstan, and down into North Africa (see map). To do this Dr Czisch used a technique called linear optimization, originally developed to solve complicated logistical problems, which is widely used in business. It took Czisch years to gather the necessary data, including detailed weather and electricity consumption data for the whole area, and investment costs for all the main renewable technologies.
Czisch then told the programme to devise the cheapest electricity supply system that could satisfy demand entirely from renewables. Based on all the data, the model would decide which forms of generation should be sited where, and plan the routes and capacity of the HVDC lines. The results were astonishing: not only could the electricity demand of more than a billion people be supplied solely from renewables throughout the year – the lights would never go out – but it also wouldn’t break the bank.
At first glance the numbers look enormous: the entire project would cost more than €1.5 trillion over 20 years, of which €128 billion would go on the Supergrid itself – the HVDC lines and equipment – and around €1.4 trillion on renewable generating capacity. But to put these figures in context, the International Energy Agency forecasts that the global power industry will have to invest $13.6 trillion by 2030 in any case, even under a business-as-usual scenario in which coal and gas continue to dominate the electricity supply. Under the Czisch plan the idea is that investment in clean technologies would displace spending on dirty ones, not add to it.
Or course wind and solar are currently more expensive than gas and coal fired plants – at least if the cost of emissions is excluded, or set unrealistically low – but one of the advantages of the Supergrid is that renewables can be sited in the best locations, where the wind and the sun are strongest and most consistent, meaning the efficiency and economics improve. And although the Supergrid itself would cost billions, because it represents only a small proportion of the total investment, the extra cost has little impact on the overall price of electricity. So even without factoring in a price for carbon, Czisch calculates the system could deliver electricity for less than 4.7 Euro cents per kilowatt hour, roughly the price of German wholesale electricity in 2005 when the study was completed.
In Czisch’s reference scenario, the bulk of the energy comes from onshore wind (70%), the cheapest form of renewable generation, with powerful summer winds in Morocco and Egypt complementing winter gales around the North Sea. Most of the rest comes from existing hydro power in the Nordic countries and the Alps, which the model holds in reserve and despatches only when the other sources fail to match demand. Another scenario shows that European demand could be satisfied entirely from renewables even without imports, but at slightly higher cost. “European politicians are still arguing about whether we can achieve our 2020 targets”, says Dr Czisch, “but my work shows there is nothing to stop us going completely renewable. It is just a question of political will”
To make the system work would mean building tens of thousands of kilometres of new HVDC lines, which in Europe could provoke public outcry and lengthy planning disputes. But Czisch argues that the Supergrid would represent only a small addition to the existing infrastructure; Germany for instance would need around 8,000 kilometres of new HVDC, whereas its existing HVAC grid spans more than 100,000km. “So for a less than 10% increase in lines we get a totally renewable electricity supply. This is not a problem; it’s a bargain”.
But public opposition could still push up the costs, as it already has with a planned HVDC interconnector between the French and Spanish grids across the eastern Pyrenees. The proposal was deadlocked for fifteen years by environmental protests, and was only approved last year when the developers finally agreed to run the cable underground. But that has raised the predicted costs from €125 million to €455-€525m. According to Alain Baron, Principal Administrator in the European Commission’s Transport and Energy Directorate, who helped broker the deal, “we have opened a Pandora’s box. If we have to bury the Supergrid over thousands of kilometres, it will cost a fortune”.
Dr Czisch insists that the France-Spain link is a special case in an environmentally sensitive area, and that most of the Supergrid would still run on cheaper overhead lines.
The HVDC Supergrid may be an ambitious vision, but in Europe potential elements are already beginning to emerge – although largely at sea – and could soon be eligible for the EU subsidies proposed in January. In the Baltic, Sweden, Germany and Denmark are investigating the potential for a three-way interconnector at Kriegers Flak, an area where all three countries plan to build offshore wind farms (see map). Since the wind farms will be close together, and each must be tied back to its respective country, connecting them with a short HVDC link could be a cheap way to allow cross-border electricity trading. According to Kim Behnke, head of R&D at Energinet, the Danish grid operator, if the idea gets the go-ahead, Kriegers Flak could be a “little Supergrid”.
In North Sea, the recently formed British company Mainstream Renewables, plans to create what it calls a ‘Supernode’, consisting of two interconnected offshore wind farms – one British, one German – with a backup connection supplying Norwegian hydro, which it hopes to complete in 2015. This demonstration project would then expand, and link to similar schemes in the Baltic, Irish Sea, Bay of Biscay and Mediterranean, to form a marine Supergrid encircling Europe (see map).
Mainstream’s founder, the Irish entrepreneur Eddie O’Connor, who recently sold his previous company, Airtricity, for €1.5 billion, is convinced that because of the difficulties of building new power lines on land, and the scale of the offshore wind resource, most of the Supergrid will be built offshore. “We could more or less do the Supergrid by sea”, he says, “and the fish won’t object”. O’Connor reckons his scheme could replace coal and gas in European power generation within thirty years.
Whether offshore or onshore, the Supergrid will depend on HVDC, but opinions differ about quite how quickly HVDC can deliver the full benefits of the Supergrid. Today almost all HVDC links are ‘point to point’, consisting of a single line or cable with one converter station at each end. But the Supergrid would need a more complicated arrangement of ‘multipoint’ links, with several converter stations along each line, each able to feed in and draw out power, along with a sophisticated control system.
Dr Czisch insists that his project could be built using only proven technology, citing an EU sponsored study into a proposed multipoint link to connect Russia, Lithuania, Poland and Germany, published in 2000, which concluded that the system was “technically feasible”. On this basis, and with the support of several retired HVDC experts, Czisch claims there are no meaningful obstacles left.
Others are more cautious. “There is a difference between what is technically possible and equipment being commercially available” says Graeme Bathurst of TNEI, “it’s far from trivial. The R&D lead times are measured in years”.
Dr Asplund of ABB says that securing the full benefits of the Supergrid would mean building not just multipoint links but a ‘meshed’ network – where if one line fails, the power automatically diverts along another to reach its destination. That would require further technology developments – such as improving the performance DC circuit breakers – but because it will take some years to build the grid, he expects those breakthroughs to come in good time.
Dr Peter Menke, Innovation Manager for the Transmission Division at Siemens, agrees with both Czisch and Asplund, but stresses the urgency of starting right away, and not waiting for further advances in the technology, especially since it can take many years to secure planning permission for new overhead lines. “The need to increase grid capacity in Europe is so great that we should start to build the HVDC backbone right now”, he says, “we can always upgrade later”.
With so much riding on HVDC, it’s likely any remaining technical problems will be cracked soon, says Graeme Bathurst: “The real issue now is not can we do it – yes we can – but do we want to”. And supporters claim the Supergrid could emerge surprisingly quickly: after all, the British national grid was built in just a decade after 1925, cutting electricity costs by a quarter. So if the politics can be squared, it looks as if Edison may soon be enjoying some posthumous revenge.