Power Electronics offers answers to demand for power
One day, buying and installing a small power plant could be as easy as ordering a custom computer. Electrical and Computer Engineering Professor Robert Lasseter and Assistant Professor Giri Venkataramanan are developing "plug-and-play" power electronic systems to simplify the configuration and installation of a variety of power generation technologies.
If every small manufacturer or power-hungry business like a hotel, restaurant or apartment building could make its own power, rolling blackouts would become a thing of the past. The natural-gas-fueled microturbine is a relatively new power-production technology that could be just the size to suit many small businesses. The mechanically simple, single-shaft generator spins between 50,000 and 100,000 rpm on airfoil bearings. It could provide reliable power in the 25 to 100 kw range. No larger than a side-by-side refrigerator, these systems have low emissions and produce heat, that when captured for heating water or other purposes, doubles the efficiency of the system.
"The advantages are several," says Lasseter. "With natural gas as a fuel, what comes out of the exhaust is very minimal compared to say, natural gas in a reciprocal engine. What happens in a reciprocal engine is the air mixing with gas creates combustion, but the nitrogen in the air ends up putting out nitric oxide. Utilities have scrubbers to keep that down but that's not a practical technology on a small scale. Microturbines are different. Their combustion temperatures are low enough that they don't form nitric oxide. So you have about one-tenth the output that you would have with a traditional natural-gas-powered generator or your home furnace." Mass production could make microturbines affordable, but installation and use would be expensive without simple, reliable power-electronics control technology that can seamlessly switch between a business making its own power and pulling power from the utility grid. Just as a utility matches the power it produces to demand from users, each small business will need "plug-and-play"-type power electronics to match the power produced to the loads that make up its own microgrid. "With microturbines, half the cost has been in the power electronic control technology," says Venkataramanan. "The design of these systems is still custom. We are trying to come up with concepts that are very similar to the digital logic industry where the design is automated and you have mass-produced components that go together and can be designed by somebody who can write computer code. They don't have to know a lot of engineering. We're coming up with standardized cell techniques that will make it easier to design, engineer and manufacture these power supplies. Buying and installing a microturbine, fuel cell or other power supply could be much like you pick components to build your own computer."
Current microgrid technology is too expensive for the restaurant down the street, but for businesses with sensitive loads, the systems could result in added revenue. A power interruption can mean millions of dollars in losses to businesses like large Internet server firms, credit card companies or electronics manufacturers. A voltage drop can cause computers to shut down and reboot or require a sensitive manufacturing process to be recalibrated. Industries like these require "clean power," or 99.9999 percent or greater reliability. Because fluctuating voltage can cost a company millions of dollars, some are willing to pay millions for superconducting energy storage systems that detect voltage fluctuations and then quickly inject power so that the loads don't see the fluctuation. "We have a whole new approach to this," says Lasseter. "What we're saying is 'We have the power right here.' If the voltage drops and we have enough power to meet all the loads, we can disconnect from the utility and then reconnect when the supply becomes stable. By putting power electronics controls on a microturbine we can deal with sensitive loads without the enormous costs of a superconducting coil and things like that." In their lab, Lasseter and Venkataramanan built a system of loads and power supplies to simulate the power requirements of a small manufacturer. After proving the concept there, they intend to seek funding for a larger demonstration project with industry.