Wider deployment of certain hardware--semiconductor-based power controllers, superconducting fault current limiters, high-voltage, direct-current lines, and the like--would make electricity grids a lot more robust [see "Five Technologies to Keep Blackouts at Bay"]. Just as important, however, are grid-management tools--both software and hardware--that have been found to work but are not used nearly as well as they might be in the United States. Such tools figured prominently in discussions at the IEEE Power Engineering Society's Transmission and Distribution meeting in Dallas the week of 8 September. Those mentioned most frequently by attendees follow.
Special protection schemes. When power outages cascade precipitously, as in the Western U.S. blackouts of July and August 1996 or the great northeastern-midwestern blackout of 14 August 2003, human operators scarcely have enough time to determine what is going on or to develop a coordinated response to keep supplies and loads balanced without overloading transmission lines or generators. Accordingly, power systems can be safer if the most likely breakdown scenarios have been identified in advance and if preprogrammed strategies have been devised to deal effectively with each scenario as soon as it's recognized--either by human operators or by computerized monitoring systems.
Such systems can be simple and local (confined to the management of individual transmission lines, for example), regional (coordinating the actions of two or more control systems), or central (providing for the security of an entire grid system, say, the Midwest's or the Northeast's). To work reliably at the highest level, protection schemes require agreed-upon communications protocols among all operating systems, redundant equipment in key monitoring systems, and provisions for coordinated responses when monitors reveal characteristic breakdown patterns.
Dynamic voltage and transient stability studies. The difficulty in developing protection schemes--especially the largest and more complex ones for an entire well-defined power grid (France's, say, or the Texas system), is that in principle an infinite number of things can go wrong. There is a premium, therefore, on credible simulation studies that determine for relevant control areas which elements of the system are most vulnerable or most critical, and what they can handle under both normal and emergency conditions. Such studies sometimes stretch mathematical and computer techniques to the utmost, and depend on more than a small element of art.
Nevertheless, if key choke points and breakdown scenarios are identified shrewdly and analyzed, the benefits can be huge. In the 14 August outage, for example, East-West trunk transmission lines passing through northeastern Ohio played a critical role in the breakdown sequence [see Timeline]. But had the two regional transmission organizations with authority over that region of the country, the PJM Interconnection and the Midwest Independent System Operator, commissioned pertinent studies, and were such studies credible? In the case of PJM, a highly experienced and well-regarded grid operator, it had discovered even before the blackout that some of its simulation studies did not accurately portray what would happen in the real world.
Smart SCADA systems. Supervisory control and data acquisition systems have been in use for many years, in power grids all around the world, to collect real-time information on system performance and to permit operators to fine-tune systems from central control rooms. Such SCADA systems also are now capable of processing as well as merely assembling data. If such capabilities were exploited more fully, operators or automated operating systems could better prevent actions taken with excessive haste to protect local equipment, without regard for how larger systems would be affected.
Wide-area measurement systems. A so-called WAMS system has been demonstrated in the Western grid system and proved its worth in the aftermath of the summer 1996 outages, when analysts used information the Western grid had collected to figure out what had happened. The real intention behind such systems, however, is to provide information to be used prospectively to prevent outages in the first place.
The general idea is to have portable system monitors distributed throughout the grid. As they collect information, each data packet is time stamped, using signals from global positioning satellites, and forwarded to central control rooms. There, the information can be processed, relying on SCADA technology described above, and evaluated in light of stability studies to order protection schemes implemented.
Transmission line monitors. On the ground, technology is being developed to monitor the state of transmission lines more closely. In both the Western and Northeastern blackouts, a surprisingly significant role was played by highly loaded lines sinking into trees, as they expanded from heating, and shorting out. Generally, thermal expansion of lines radically limits the amount of power they can carry at any given time (and new materials for lines are being developed that will let them expand much less and therefore carry much more power).
Southwire Co. (Carrollton, Ga.) has tested an electricity cable in which fiber-optic material is embedded in the entire length of the line, so that its status can be checked continuously. Oncor, a transmission and distribution subsidiary of TXU (Dallas), exhibited a system it calls CAT-1 at the Dallas Transmission and Distribution meeting: it computes how much a line is sagging from data taken at each transmission tower.
An especially comprehensive account of the techniques that need to be used more systematically to prevent blackouts was presented at a special session on the 14 August blackout, held Tuesday morning on 9 September in Dallas, by IEEE Fellow Damir Novesel, general manager of KEMA T&D Consulting (Chalfont, Pa.). The Powerpoints of that talk, and of the other four talks delivered in that session, are to be posted on the IEEE Power Engineer Society's Web site at: