What gets the most attention is not what causes blackouts in North America and Europe. It’s the system, not a shortage of power plants that is the problem. Take a look at the 13 major power outages over many years, and see that the problems we face are not because we aren’t building enough power plants. Part one of a two-part series on the Northeast Blackout of 2003.
Only one of these outages, July 2012 in India, was due to more electricity demand than could be supplied by existing resources. In the industrialized economies of North America and Europe, we more often lose power due to a subtle and difficult challenge. The electrical grid is prone to system failures and needs modernization.
Crews work on line. Credit: Mike Jacobs
For decades the concern over power grid reliability focused on ensuring that an adequate number of power plants were built. Still today, most of the policy attention, the financial needs, and advanced planning are on building enormous new plants. This is a holdover from past decades when growth in electricity use was high, and the time it took to build a power plant was increasing. But when one looks at what has caused major blackouts, insufficient power plants was only a factor in the India example, where people are being added to the Age of Electricity as services gradually reach more communities.
In North America and Europe, we have a different set of concerns. Load growth is barely 1 percent per year and there have been significant investments in new generation and technologies to save energy and use renewable energy. Still, every year the regulators and the utility industry make a number of announcements comparing the expected demand and the expected supply. In many states, this reporting is required by law. The numbers in these comparisons are easy math. When reviewed, everyone feels assured that the power supply is large enough to meet demand, or that the investments are coming and the required bills for this assurance will be paid. Even Texas, with its energy crunch, has 150 new plants in the planning process.
Unfortunately, it is unexpected disturbances, usually on the wires, that cause almost every blackout. Storms, droughts, and fires knock out whole sections of the system; control errors and flubbed operations trigger shutdowns; coordination failures cause overloads. Transmission reliability is much more complex than the adequacy of the generation fleet.
The August 2003 Northeast Blackout resulted from a combination of key monitoring systems offline, generators not responding as anticipated or requested, and then an overloaded line sagging from excess heat and short-circuiting to a tree. Obvious to the experts, this blackout could have been prevented if the grid reliability rules, including tree trimming, were mandatory, and the system needs for communications and cooperation were enforceable.
While the attention of utilities and politicians has been on the largest power plants, the practices for running the system were neglected. Coordination between utilities, adoption of flexible schedules, and use of accurate forecasts allow the transmission system to work reliably. Responsibility had been divided by old territorial boundaries between utility companies, even as the system was becoming more regional.
The creation and strengthening of the regional Independent System Operators has brought great progress inside the regions these serve. However, the utility industry continues to struggle to improve power flows across boundaries, information sharing, and cooperation. These reforms are vital to increasing reliability and lowering costs. We will see in the next post in this series that this modernization will help integrate wind and solar energy supplies with the rest of the grid.
In the summary of 13 power outages below, notice how the weather and the operations of the grid caused the blackouts. Coordination and better information, rather than more old-fashioned power plants, are the recurring need for more reliable systems.
1) October 2012 Hurricane Sandy: Flooding damaged vulnerable equipment and downed trees cut power to 8.2 million people in 17 states, the District of Columbia, and Canada, many for 2 weeks. The impacts from sea level rise and flooding are leading to re-evaluation of local design criteria.
2) July 30 and 31, 2012 Northern India: High demand, inadequate supply coordination, and transmission outages led to a repeating power system collapse that affected hundreds of millions across an area home to half of India’s population. Four key transmission lines were taken offline in previous days. Mid-summer demand in the north exceeded local supply, making the imports and transfers from west vital. Excessive demand tripped a transmission line. Within seconds, ten additional transmission lines tripped. Conditions and failure repeated again the following day. A review found poor coordination of outages and regional support agreements.
3) June 2012 Derecho: Wind storm damaged trees and equipment, cutting power to approximately 4.2 million customers across 11 Midwest and Mid-Atlantic states and the District of Columbia. Widespread tree clearing and line restoration efforts in many cases took 7 to 10 days.
4) October 2011 Northeast U.S.: Record early snowstorm brought down trees and wires. Outages could only follow removal of snow and fallen trees. More than three million customers in Mid-Atlantic and New England states were without power, many over 10 days.
5) September 8, 2011 California-Arizona: Transmission failure was set up by Southern California’s heavy dependence on power imports from Arizona, an ongoing problem. Hot weather after the end of the summer season, as determined by the engineering schedule, conflicted with generation and transmission outages planned for maintenance. Then two weaknesses — operations planning and real-time situational awareness — left operators vulnerable to a technician’s mistake switching major equipment. This outage lasted 12 hours, affecting 2.7 million people.
6) August 28, 2003 London: Two cables failed, and a leaky transformer could not handle the resulting flows. A section of the city and southern suburbs, totaling 250,000 customers, were off from 6:30 to 7 pm when alternate circuits were arranged.
7) August 14, 2003 Northeastern US and Ontario: Transmission system failed for many reasons seen in major outages that came years before. Information was incomplete and misunderstood; inadequate tree trimming caused short circuit; operators lacked coordination. System imbalances and overloads seen early in the day were not corrected due to lack of enforcement of coordination. 50 million people across eight states and Ontario were without power up to four days.
8) June 25, 1998 Ontario and North Central U.S.: A lightning storm in Minnesota initiated a transmission failure. A 345-kV line was struck by lightning. Underlying lower voltage lines overloaded. Soon, lightning struck a second 345-kV line. Cascading transmission line disconnections continued until the entire northern Midwest was separated from the Eastern grid, forming three isolated “islands” with power. 52,000 people in upper Midwest, Ontario, Manitoba, and Saskatchewan saw outages up to 19 hours.
9) July 2-3, 1996 West Coast: The transmission outage began when a 345-kV line in Idaho overheated and sagged into a tree. A protective device on a parallel transmission line incorrectly tripped that line. Other relays tripped two Wyoming coal plants. For 23 seconds the system remained in precarious balance, until a 230-kV line between Montana and Idaho tripped. Remedial action separated the system into five pre-engineered islands to minimize customer outages. Two million people in the U.S., Canada, and Mexico lost power for minutes to hours.
10) August 10, 1996 West Coast: Hot weather and inadequate tree trimming set up transmission collapse. Through the afternoon five power lines in Oregon and nearby Washington short-circuited on trees. This tripped off 13 hydro turbines operated by BPA at McNary Dam on the Columbia River. Blame fell on inadequate tree-trimming practices, operating studies, and instructions to dispatchers. Approximately 7.5 million customers lost power in seven western U.S. states, two Canadian provinces, and Baja California, Mexico for periods ranging from several minutes to six hours.
11) December 22, 1982 West Coast: Over 5 million people in the West lost power after high winds knocked over a major 500-kV transmission tower. The tower fell into a parallel 500-kV line tower, and the failure mechanically cascaded and caused three additional towers to fail on each line. When these fell, they hit two 230-kV lines crossing under the 500-kV lines. From that point, coordination schemes did not operate, communication problems delayed control instructions. Backup plans failed because the coordination devices were not set for such a severe disturbance. Data displayed to operators was unclear, preventing corrective actions.
12) July 13, 1977 New York City: Transmission failures caused by lightning strike shutting lines, and the tripping offline Indian Point No. 3 nuclear generating plant. When a second lightning strike caused the loss of two more 345-kV lines, the last connection for New York City to the northwest was lost. Power surges, overloads, and human error soon followed. Nine million people in New York City suffered outages and looting up to 26 hours. Poor coordination, malfunctioning safety equipment, and limited awareness of conditions contributed to the outage.
13) November 9, 1965 Northeast U.S. and Ontario: Transmission system failed due to a mistaken setting on a protective device near Niagara Falls. Improper coordination caused four more lines to disconnect. Imbalances continued to swing until power was out for 30 million people. The outage lasted up to 13 hours.