Categories
Select Country
Follow Us
Supercomputer

This Software Can Take the Heat: Stanford Spinoff is Helping GE Develop Tomorrow’s Turbines

She’s a massive beast that can generate up to 600 megawatts of electricity in a combined cycle power plant, the equivalent power that would be needed to supply approximately 600,000 U.S. homes. To do so efficiently, however, this latest GE gas turbine, officially called 9HA but nicknamed Harriet by GE workers, must withstand temperatures greater than 2,600 degrees Fahrenheit.

But here’s the rub: over the long term, the high heat also causes wear and tear inside the turbine that is impossible to see. Impossible, that is, without a virtual model of the turbine operating at full capacity.

image

Cascade is using powerful software to model turbulent combustion. The program allows the company to monitor and track pollutants like CO, CO2 and NOx. GIF credits: Cascade Technologies

That’s why GE started working with Cascade Technologies, maker of powerful simulation software, to study complex fluid and heat flows inside big machines like gas turbines.

The relationship with Cascade, a spin-off from the Center for Turbulence Research at Stanford University, will give GE engineers a virtual peek inside Harriet, the world’s largest and most efficient gas turbine, and help them build even more efficient machines in the future.

image

Frank Ham, Cascade president and CEO, says that the software is the equivalent of a “modern-day digital microscope.”

“Seeing details they were not aware of helps engineers better understand why gas turbine designs work the way they do, and GE gains critical knowledge as to how they can improve them,” Ham says.

The digital microscope that Ham is talking about is powerful software that can run on some of the world’s most powerful computers, including supercomputers operated by U.S. national labs. The software is able to process petabytes of data, which is roughly four times the amount of information held by the U.S. Library of Congress. (GE Aviation is also using supercomputers to design more efficient jet engines.)

image

All of that data can yield powerful simulations of a number of scenarios, including the multiple step combustion process inside a turbine. The code is so detailed that it replicates this intricate system at the microsecond level.

All of that computing power should allow GE engineers to make changes and improvements to new turbine designs up to ten times faster over the course of a typical two-year product development process. That means more robust, efficient and cleaner turbines, delivered faster, according to John Lammas, vice president of power generation engineering at GE Power & Water.

Says Cascade’s Ham: “By providing information that goes into design decisions, we can help improve efficiency, lower emissions and increase durability in future products.”

image

Subscribe to our GE Brief