Rise of the supergrid
Electricity now flows across continents, courtesy of direct current
Transmitting power over thousands of kilometres requires a new electricity infrastructure
THE winds of the Oklahoma panhandle have a bad reputation. In the 1930s they whipped its over-tilled topsoil up into the billowing black blizzards of the Dust Bowl. The winds drove people, Steinbeck’s dispossessed, away from their livelihoods and west, to California.
Today, the panhandle’s steady winds are a force for creation, not destruction. Wind turbines can generate electricity from them at rock-bottom prices. Unfortunately, the local electrical grid does not serve enough people to match this potential supply. The towns and cities which could use it are far away.
So Oklahoma’s wind electricity is to be exported. Later this year, lawsuits permitting, work will begin on a special cable, 1,100km (700 miles) long, between the panhandle and the western tip of Tennessee. There, it will connect with the Tennessee Valley Authority and its 9m electricity customers. The Plains and Eastern Line, as it is to be known, will carry 4,000MW. That is almost enough electricity to power Greater London. It will do so using direct current (DC), rather than the alternating current (AC) that electricity grids usually employ. And it will run at a higher voltage than such grids use—600,000 volts, rather than 400,000.
This long-distance ultra-high-voltage direct-current (UHVDC) connector will be the first of its kind in America. But the problem it helps with is pressing everywhere. Fossil fuels can be carried to power stations far from mines and wells, if necessary, but where wind, solar and hydroelectric power are generated is not negotiable. And even though fossil fuels can be moved, doing so is not desirable.
Coal, in particular, is costly to transport. It is better to burn it at the pithead and transport the electricity thus generated instead.
Transmitting power over thousands of kilometres, though, requires a different sort of technology from the AC now used to transmit it tens or hundreds of kilometres through local grids. And in China, Europe and Brazil, as well as in Oklahoma, a new kind of electrical infrastructure is being built to do this. Some refer to the results as DC “supergrids”.
When it flows as direct current, there is no oscillation. Both work well, but the deciding factor in AC’s favour in the 19th century was the transformer. This allows AC voltages to be increased after generation, for more efficient transmission over longish distances, and then decreased again at the other end of the line, to supply customers’ homes and businesses. At the time, direct current had had no such breakthrough.
When one eventually came, in the 1920s, in the form of the mercury arc valve, AC was entrenched.
Even the solid-state thyristor, a cousin of the transistor invented in the 1950s, offered no great advantages over the tens or hundreds of kilometres that power grids tended to span. Some high-voltage DC lines were built, such as that under the English Channel, linking Britain and France. But these were justified by special circumstances. In the case of the Channel link, for example, running an AC line through water creates electromagnetic interactions that dissipate a lot of power.
Over transcontinental distances the balance of advantage shifts. As voltages go up, to push the current farther, AC employs (and thus wastes) an ever-increasing amount of energy in the task of squeezing its alternations through the line. Direct current does not have this problem. Long-distance DC electrical lines are also cheaper to build. In particular, the footprint of their pylons is smaller, because each DC cable can carry far more power than an equivalent AC cable.
Admittedly, thyristors are expensive—the thyristor-packed converter stations that raise and lower the voltage of the Plains and Eastern line will cost about $1bn, which is two-fifths of the project’s total bill. But the ultra-high voltages required for transcontinental transmission are still best achieved with direct current.
For all the excitement surrounding the Plains and Eastern Line, however, America is a Johnny-come-lately to the world of UHVDC. Asian countries are way ahead—China in particular. As the map on the previous page shows, the construction of UHVDC lines is booming there. That boom is driven by geography. Three-quarters of China’s coal is in the far north and north-west of the country. Four-fifths of its hydroelectric power is in the south-west. Most of the country’s people, though, are in the east, 2,000km or more from these sources of energy.
China’s use of UHVDC began in 2010, with the completion of an 800,000-volt line from Xiangjiaba dam, in Yunnan province, to Shanghai. This has a capacity of 6,400MW (equivalent to the average power consumption of Romania). The Jinping-Sunan line, completed in 2013, carries 7,200MW from hydroelectric plants on the Yalong river in Sichuan province to Jiangsu province on the coast. The largest connector under construction, the Changji-Guquan link, will carry 12,000MW (half the average power use of Spain) over 3,400km, from the coal- and wind-rich region of Xinjiang, in the far north-west, to Anhui province in the east. This journey is so long that it requires 1.1m volts to push the current to its destination.
China’s UHVDC boom has been so successful that State Grid, the country’s monopolistic electricity utility, which is behind it, has started building elsewhere. In 2015 State Grid won a contract to build a 2,500km line in Brazil, from the Belo Monte hydropower plant on the Xingu River, a tributary of the Amazon, to Rio de Janeiro.
China’s neighbour India is following suit—though its lines are being built by European and American companies, namely ABB, Siemens and General Electric. The 1,700km North-East Agra link carries hydroelectric power from Assam to Uttar Pradesh, one of the country’s most densely populated areas. When finished, and operating at peak capacity, it will transmit 6,000MW. At existing levels of demand, that is enough for 90m Indians. The country’s other line, also 6,000MW, carries electricity 1,400km from coal-fired power stations near Champa, in Chhattisgarh, to Kurukshetra, in Haryana, passing Delhi on the way.
This tendency to fan out is another reason it is hard to corral power over long distances through AC grids—for, being grids, they are made of multiple, interconnected lines. Despite UHVDC connectors being referred to as supergrids, they are rarely actual networks. Rather, they tend to be point-to-point links, from which fanning out is impossible. Some utilities are therefore looking at them to move power over relatively short distances, as well as longer ones.
One such is 50Hertz, which operates the grid in north-east Germany. Almost half the power it ships comes from renewable sources, particularly wind. The firm would like to send much of this to Germany’s populous south, and on into Austria, but any extra power it puts into its own grid ends up spreading into the neighbouring Polish and Czech grids—to the annoyance of everyone.
50Hertz is getting around this with a new UHVDC line, commissioned in partnership with Germany’s other grid operators. This line, SuedOstLink, will plug into the Meitingen substation in Bavaria, replacing the power from decommissioned south-German nuclear plants. And Boris Schucht, 50Hertz’s boss, has bigger plans than that. He says that within ten years UHVDC will stretch from the north of Sweden down to Bavaria. After this, he foresees the development of a true UHVDC grid in Europe—one in which the lines actually interconnect with each other.
That will require new technology—special circuit-breakers to isolate faulty cables, and new switch gear—to manage flows of current that are not simply running from A to B. But, if it can be achieved, it would make the use of renewable-energy sources much easier. When the wind blows strongly in Germany, but there is little demand for the electricity thus produced (at night, for instance), UHVDC lines could send it to Scandinavian hydroelectric plants, to pump water uphill above the turbines.
That will store the electricity as potential energy, ready to be released when needed. Just as sources of renewable energy are often inconveniently located, so, too are the best energy-storage facilities.
UHVDC permits generators and stores to be wired together, creating a network of renewable resources and hydroelectric “batteries”.
In Asia, something similar may emerge on a grander scale. State Grid plans to have 23 point-to-point UHVDC links operating by 2030. But it wants to go bigger. In March 2016 it signed a memorandum of understanding with a Russian firm, Rosseti, a Japanese one, SoftBank, and a Korean one, KEPCO, agreeing to the long-term development of an Asian supergrid designed to move electricity from windswept Siberia to the megalopolis of Seoul.
This project is reminiscent of a failed European one, Desertec, that had similar goals. But Desertec started from the top down, with the grand vision of exporting the Sahara’s near-limitless solar-power supply to Europe. Today’s ideas for Asian and European supergrids are driven by the real needs of grid operators.
Such projects—which are transnational as well as transcontinental—carry risks beyond the merely technological. To outsource a significant proportion of your electricity generation to a neighbour is to invest huge trust in that neighbour’s political stability and good faith. The lack of such trust was, indeed, one reason Desertec failed. But if trust can be established, the benefits would be great.
Earth’s wind-blasted and sun-scorched deserts can, if suitably wired up, provide humanity with a lot of clean, cheap power. The technology to do so is there. Whether the political will exists is the question.