veski innovation fellow Richard Sandberg awarded an INCITE project

Principal Investigator and veski innovation fellow Professor Richard Sandberg and Co-Investigator Yaomin Zhao have been awarded an INCITE project through the US Department of Energy (DOE) Office of Science, who provide a portfolio of national high-performance computing facilities housing some of the world’s most advanced supercomputers. These leadership computing facilities enable world-class research for significant advances in science. Along with three additional co-investigators from General Electric Aviation, they were awarded 525,000 node hours on IBM AC922 (Summit) at the Oak Ridge National Laboratory.

The award of 525,000 node hours on the fastest supercomputer in the world is equivalent to more than the entire merit allocation in Australia (for all researchers in all fields) and worth approximately $14 million if it was on Australian machines, with Summit having a peak performance of 200,000 trillion calculations per second.

Gas turbines are—and will continue to be—the backbone of aircraft propulsion, power generation, and mechanical drive due to their power density (i.e., thrust per unit engine weight), efficiency, and ability to adjust to rapidly varying loads. In the US alone, the natural gas and oil burn summed up to 27 × 1012 cubic feet and 6.3 × 109 barrels of oil equivalent respectively in 2015. Therefore, even at the current fuel price, a small engine performance improvement does have a fuel-spend advantage of the billion-dollar order, together with a significant CO2 emission benefit.

The Melbourne/GE team is exploiting the capability of the very efficient computational fluid dynamics code, the High-Performance Solver for Turbulence and Aeroacoustic Research (HiPSTAR) developed by Sandberg’s research group, to perform the first-of-a-kind direct numerical simulation of high-pressure turbine stages with realistic geometry and at engine-relevant conditions.

The generated data will shed light on the detailed fundamental flow physics—in particular the behavior of transitional and turbulent boundary layers affected by large-scale violent freestream turbulence—under strong pressure gradient and curvature. It will also help evaluate and develop lower-order models readily applicable to gas turbine designs. With the results, it will be possible to identify opportunities to increase turbine aerothermal efficiency by 2–4 percent and extend hot-gas-path durability. This would translate into combined cycle efficiency gains of 0.4–0.8 percent and thus have a significant economic and environmental impact.

Courtesy of the University of Melbourne & Oakridge National Laboratory US.