TELOS about an outage

A team of engineers led by a Purdue University researcher plan to demonstrate a new system that aims to avert power failures by automatically adapting to the daily fluctuations in electricity consumption.

If successful the system, scheduled for demonstration in April, might be helpful in easing problems associated with electricity shortages and utility deregulation in the US.

The system, dubbed for Transmission Entities with Learning Capabilities and On-Line Self-Healing (TELOS) will predict and monitor electricity consumption for each customer on the hour and make it possible to charge higher rates for those placing the greatest strain on the power grid during times of peak demand.

Computerised systems are currently capable of gauging how much electricity a customer is using from one hour to the next. Some customers consume power in predictable patterns, which is said to make it easier for utilities to manage electrical distribution. Other customers are less predictable, complicating efforts to keep the system from crashing.

TELOS is said to be a ‘self-healing’ system because it automatically adjusts to new conditions. It is designed to prevent power failures by anticipating the quickly changing demands of industrial, commercial and residential electric customers.

The system has been designed to accurately predict power needs for the coming day and then automatically meet those demands by better managing electricity distribution and supplementing the grid with power from small natural gas or diesel generators, which would operate when needed.

‘We want to endow the grid with certain self-regulation capabilities,’ said Lefteri Tsoukalas a professor of nuclear engineering at Purdue.

Such an automated system would enable sophisticated control of small sections within small service areas. These small sections would be called local area grids (LAGs).

During times of peak demand, TELOS would combine LAGs so that customers in one area who were expected to increase their consumption over the next hour might be offset by customers in another LAG who were expected to reduce their consumption within the same period. The predictions would be made using computer databases containing a history of each customer’s electrical consumption.

Presently, damage to a single substation in the grid can trigger a domino effect, where a series of equipment failures leads to a widespread power failure. Isolating problems to a particular LAG would, according the Purdue team, prevent widespread power failures.

Research findings show that TELOS would be effective in controlling the flow of electricity during times of peak demand, such as a summer heat wave.

TELOS is scheduled to be completed by the end of 2001 and to be operating in 2003 on a trial basis in the Commonwealth Edison and Tennessee Valley Authority service areas of the US. If it works, TELOS will be available for use on a national level, concluded Tsoukalas.