In the not-too-distant-future, the U.S. will join the growing community of nations that save money and energy by transporting electricity through direct current (DC) rather than alternating current (AC) transmission lines. The 700-mile-long Plains & Eastern Clean Line will supply the 9 million customers of the Tennessee Valley Authority with 3,500 megawatts (MW) of electric power from the wind farms of the Oklahoma panhandle. Operating at 600 kilovolts instead of the usual 400, the high-voltage direct current (HVDC) line will also supply Pope County, Ark., with an additional 500 MW.
HVDC transmission line losses are quoted at less than three percent per 1,000 km, some 30-40 percent lower than AC lines operating at similar voltages. The line-loss advantage is even greater for ultra-high-voltage direct current (UHVDC) lines. Neither advantage is high enough to justify the use of HVDC or UHVDC lines over traditional distances, but both are well worth exploiting for the longer distances that wind and solar power will soon need to travel to link the sunniest and windiest parts of the planet with existing population centers. Long-distance DC lines are also cheaper to build than AC lines because DC cables can carry significantly more power than AC cables of equal cross section — largely due to something known as the “skin effect.” Among other things, this permits reduction of the supporting pylons’ footprint.
Figure 1. Illustration of the skin effect in the cross-section of an alternating current (AC) transmission line. Image credit: James Case.
“Skin effect” refers to the fact that the distribution of AC in a cylindrical conductor is far from uniform, as indicated in Figure 1. In reality, the current density decreases exponentially from the surface toward the central axis. The “skin depth” \(\delta\) is defined as the depth at which the current density is just \(1/e\) (about 37 percent) of the value at the surface. This varies from one installation to another, depending on the frequency of the current and the electromagnetic properties of the conductor itself.
Like most other HVDC and UHVDC lines, the Plains & Eastern Clean Line will accept low-voltage AC from nearby wind farms at the Pioneer Sky Energy Center outside Guymon, Okla., step it up to the required voltage via conventional transformers, convert it to DC, and ship it (in this case) eastward. At the Delta Landing Energy Center in Millington, Tenn.—not far from Memphis—the process will be reversed. Arriving high-voltage DC will be “inverted” into AC, stepped down to conventional voltages, and injected into the local power grid for use by existing customers. In addition, the Traveler Junction Energy Center in central Arkansas will siphon off and invert roughly an eighth of the power, delivering it for use in nearby Oklahoma. GE Energy Connections has been chosen to build the converter stations for the Plains & Eastern Clean Line.
To transform AC to DC and back, the energy centers will use banks of thyristors, similar to the ones long used as logic gates in electronic circuitry before finding application in the power transmission industry. Thyristors differ from transistors in that they comprise four (rather than two) alternating blocks of N- and P-type semiconducting material. They have an anode A at one end, a cathode C at the other, and a gate connected by wire to the P block nearest the cathode (see Figure 2).
Figure 2. Schematic diagram of a thyristor. Image credit: James Case.
Thyristors can be at rest in either an “open” (conducting) or “closed” (nonconducting) state, and can toggle back and forth between the two states using signals received via the gate wire. Once placed in a particular state, they tend to remain “locked” in that state until switched to the other by a signal. Because thyristors are individually small, entire buildings of them are required to rectify and invert industrial-strength currents. They are also expensive; the converter stations along the Plains & Eastern Clean Line will cost approximately one billion dollars, roughly 40 percent of the entire project’s budget.
Although thyristor installations can both rectify and invert electrical currents, most HVDC and UHVDC power lines are optimized—as is the Plains & Eastern Clean Line—for transmission in one direction only. The HVDC line beneath the English Channel, intended to send excess U.K. production to France and vice versa, is an exception. Despite the short distance separating the two nations, DC transmission is economical because AC line losses are exaggerated by underwater transmission. For similar reasons, several other short-haul underwater DC transmission lines—such as the Sardinia-Italian mainland line, which also provides power to the island of Corsica—operate in parts of Europe. DC transmission can also connect incompatible grids, such as one operating at 50 hertz (as is common in much of Europe) with another operating at 60 hertz.
The U.S. will hardly be the first country to reap the benefits of DC transmission over long distances. China began its exploitation of UHVDC technology in 2010 with the completion of an 800-kilovolt line—with 6,400-megawatt capacity—from the Xiangjiaba Dam in remote Yunnan province to Shanghai. The Jinping-Sunan line, completed in 2013, carries 7,200 MW of power from hydroelectric plants along the Yalong River in Sichuan province to coastal Jiangsu province. Presently under construction is a 12,000 MW line from coal and wind-rich regions in the far northwest to Anhui province, 2,000 miles away in the east. While 75 percent of China’s coal is located in the far north and west of the country and 80 percent of its hydroelectric power is in the southwest, most of the country’s vast population is in the east, thousands of miles from these abundant sources of energy.
So successful is China’s venture into UHVDC technology that the State Grid Corporation of China, the country’s monopolistic electricity utility, has begun building elsewhere. In 2015, it won a contract to build a 1,500-mile line in Brazil, from the Belo Monte hydropower plant on an Andean tributary of the Amazon to Rio de Janeiro.
India is also climbing aboard the UHVDC bandwagon. When finished and operating at full capacity, its 1,000-mile North East-Agra line will transmit 6,000 MW from Assam to Uttar Pradesh, one of the country’s most populous regions. A second line of the same capacity will carry power 875 miles from power plants near coal fields in the northwest, past an intermediate inversion station near New Delhi, to end users in the east. Siemens, General Electric, and the Swedish-Swiss multinational Asea Brown Boveri—rather than State Grid—are building the Indian lines.
50Hertz, the firm that operates the electric power grid in northeast Germany, is currently planning to exploit the point-to-point nature of DC transmission lines. Almost half of the firm’s generated power comes from renewable sources, particularly wind. It would like to send much of that power to southern Germany and on into Austria. But any extra power it puts into its own grid ends up energizing the Polish and Czech grids on its way to Bavaria, irritating all those involved.
The proposed solution is a new UHVDC line, known as the SuedOstLink, extending from the shores of the North Sea to an inversion station in Bavaria. Within 10 years, UHVDC lines could stretch from the far north of Sweden to Bavaria, Austria, and much of central Europe. Their creation could even lead to a true UHVDC grid in Europe, in which DC transmission lines connect with one another as well as AC suppliers and servers.
In Asia, an even more ambitious plan is afoot. State Grid intends to have 23 point-to-point UHVDC links in operation by 2030, and is looking for additional opportunities. In March 2016, the company signed a memorandum of understanding with the Russian firm Rosseti, the Japanese firm SoftBank, and the Korea Electronic Power Corporation for the development of a system to move electric power from windswept Siberia to populous Seoul.
Such projects—which are transnational as well as transcontinental—are not without risk. A country that outsources a significant proportion of its electricity generation to another nation invests great trust in the latter’s political stability and good faith. A lack thereof appears primarily responsible for the failure of an enterprise called Desertec, intended to connect the Sahara’s all-but-limitless potential for solar power generation with European markets.
If and when built, such installations will qualify as “supergrids,” where HVDC and UHVDC trunk lines connect AC grids and one another across natural barriers and national boundaries alike. The command and control of such networks, to account for the day-to-day and even hour-to-hour fluctuations of supply and demand, will be both complex and critical. While the physical know-how to construct such networks already exists, the managerial know-how to keep them operating efficiently over long periods of time has yet to be developed.