Smart Controller Makes Clean Decisions

The software is designed to work with the Gas Research Institute’s Fuel Lean Gas Reburn technology, a process that injects controlled amounts of natural gas into the upper furnaces of coal-fueled boilers. The natural gas reburns the NOx gases, converting them to elemental nitrogen, which makes up about 80 percent of the earth’s natural atmosphere.

Argonne engineers, working with industrial partner Energy Systems Associates (ESA), developed a new controller that advises power plant operators when NOx emissions are too high and suggests the appropriate degree of Lean Gas Reburn adjustments to reduce them.

“Since natural gas is more expensive than coal,” said Jaques Reifman, the Argonne project leader, “we’re interested in developing a control system that will help plant operators use natural gas economically while maintaining the required NOx reductions.”

With the controller, utilities can more easily balance the price of gas against the expense of buying emissions allowances—which can cost about $2,000 or more per ton of NOx during the ozone season.

“Initial simulation tests on coal-fueled boilers showed that the new controller consistently leads to greater NOx reduction and lower average rates of natural gas consumption when compared with manual control,” said Reifman.

The intelligent controller learns the power plant behavior using artificial intelligence methods, such as neural networks and fuzzy logic. Neural networks learn like a human does, by trial and error. The controller learned to mimic the complex plant processes in the boiler by analyzing performance data collected from a ComEd power plant.

“The plant performance database includes the amount and distribution of natural gas and the resulting NOx levels,” Reifman said. When installed in a coal-fired power plant, the NOx controller is projected by Reifman to pay for itself within one year.

For more information please contact Dave Baurac at 630-252-5584

Next: Surface Studies of Minerals Reveal Insight into Cleaning Contaminated Environments