(John Kemp is a Reuters market analyst. The views expressed are his own)
By John Kemp
LONDON, Nov 22 (Reuters) - Businesses and households expect reliable electricity to be available at the flick of a switch. But the frequency of large-scale blackouts in the United States has not fallen in the last 30 years, and big blackouts remain a common occurrence throughout emerging markets.
The risk of massive failures affecting the supply to millions of customers and darkening whole regions or countries may even be increasing as demand rises, especially for air conditioners, and electricity networks are integrated over increasingly wide areas.
India’s power failure earlier this year has heightened concern about the reliability of power networks integrated over wide areas, and raised questions about the wisdom of China’s plan for an ultra-high voltage super-grid.
On July 31, power cuts rolled across 22 states in the north of India, which are home to 620 million people (and about 320 million electric customers), about 9 percent of the global population. It was easily the biggest power outage in history.
Other mass blackouts recently have included the Java-Bali blackout in Indonesia in 2005 (100 million people); the 2009 Brazil-Paraguay blackout (which left the whole of Paraguay and parts of Brazil without electricity); and the 1999 South Brazil blackout (75 million people).
But the advanced industrial countries are not immune -- though widespread failures are much rarer. In August 2003, a blackout cut power to 50 million people across the Northeast United States and neighbouring parts of Canada, in some cases for up to four days.
Just a month later, in September 2003, a blackout cut power to 4 million people across southern Sweden and eastern Denmark. Five days after that, the whole of Italy was plunged into darkness by a cascading power failure across the country’s grid.
Blackouts are expensive. The August 2003 blackout in the United States resulted in $3 billion of insurance claims, according to one estimate.
But many insurance policies exclude power failure, and in any event only cover the direct economic costs, not the inconvenience and widespread disruption to daily life and business activity. The total cost of the August 2003 blackout was almost certainly many times the published insurance claims.
Restarting a network after a big blackout is no simple task. Most power plants rely on power from the grid in their start up procedure to work fuel and pumping systems and other control equipment. Only a small proportion of power plants are equipped with “black start” capability to begin generating power again in the event the grid is lost completely.
In a sign that should worry policymakers, the frequency of big blackouts does not appear to be falling despite heavy investment on power transmission networks and improvements in grid control systems.
In the United States, the frequency of major blackouts (supply interruptions involving at least 300 megawatts or 50,000 customers) does not appear to have decreased between 1984 and 2006, according to one recent analysis of large-scale interruptions (“Trends in the History of Large Blackouts in the United States” 2008).
In fact, the risk may actually be increasing as a result of changes in modern power system. Studies show the risk of failure increases as demand on the grid rises and the margin of spare capacity shrinks. In the United States, big power failures are most likely to occur in late afternoon on hot summer or cold winter days when air conditioning or heating demand is greatest.
Across much of the developing world, capacity margins are shrinking as investment in generation and transmission struggle to keep pace with burgeoning power demand. Low levels of spare capacity contributed to India’s mega-blackout this summer.
But margins are also shrinking in advanced economies as coal-fired power stations are taken out of service and market-based electricity systems fail to provide big enough incentives to maintain large amounts of back up capacity.
In an further wrinkle, much of the growth in demand is for airconditioning. There are concerns about the way in which the modern high-efficiency motors used in many airconditioners interact with the grid.
Many are exceptionally sensitive to a very brief drop in voltage on the network and react in ways that can make the voltage drop even worse -- known as microvoltage collapse.
“There is concern that these microvoltage collapse events will begin to lump or interact together and cause large-scale voltage collapse,” the U.S. Oak Ridge National Laboratory wrote in a 2008 report (“Local Dynamic Reactive Power for Correction of System Voltage Problems”).
In August 2003, it was a local voltage collapse in the Cleveland-Akron area of Ohio, on a summer day, that helped pull down the power supply to the entire U.S. Northeast .
“These levels of airconditioning motor load truly present a concern, because if several circuits stall at the same time, the resulting level of inductive current flow may cascade into a wider-scale voltage collapse,” Oakridge warned.
Power supplies are being networked over wider and wider areas, increasingly the risk a local problem will cascade across the entire network and cut power to millions at a time.
The July 2012 blackout rolled across India’s four northern grids, which have been linked and synchronised in recent years, highlighting the risks as well as benefits of interconnection. The fifth grid, serving the south of the country, which is not due to be synchronised until 2014, and was still relatively isolated remained unaffected.
So the current fashion for connecting up local and regional power networks to form national or even cross-national super-grids may be unwitting making them more vulnerable to widespread failures.
China is a case in point. The country is building the biggest super-grid of them all. State Grid Corporation is lobbying the government to approve as many 20 ultra-high voltage power lines criss-crossing the country, to help move power from generators in the coal-rich north and northwest, and hydropower from the gorges of Yunnan and Sichuan, to the main consuming centres along the south and east coasts.
But critics warn it could also increase the risk of local problems cascading across the national network and plunging the entire country into darkness, as happened in India.
“The building of a large-scale long-distance ultra-high voltage transmission system should be avoided,” the Energy Research Institute wrote in a recent report to the government’s powerful National Development and Reform Commission (“China Energy Outlook” 2012).
“This kind of system is relatively fragile ... If all the country’s grids are connected by ultra-high voltage lines, if the lines broke because of some natural or human incident, it would result in a national power outage. The potential hazard is huge.”
In any event, some aspects of reliability, such as local voltage collapse, are much easier to handle if there is plenty of generation in the affected area, rather than relying on transmission from distant generating units hundreds or even thousands of kilometres away.
The August 2003 blackouts proved a wake-up call for policymakers and grid companies; a lot of work has since been done to improve grid management procedures and staff training.
The focus is likely to be redoubled following India’s dramatic problems this summer. China’s State Grid has already held a special conference to examine what lessons can be learned.
But the truth is that the risk of major blackouts affecting millions or hundreds of millions of customers at a time is probably still increasing, as the rush to connect up networks outstrips in improvements in control. (Editing by William Hardy)