$17 million grant to advance simulations of hypersonic flight
BY DAVID ORENSTEIN
The promise of an airplane capable of cruising at hypersonic speeds is exhilarating: traveling from San Francisco to Sydney in 90 minutes, or taking off from an ordinary airport and flying into orbit around the Earth.
An important reason why such technology is still experimental, however, is that hypersonic vehicles are incredibly expensive and risky to build and test. As a result, accurate computer simulation is critical. To fill that need, a new $17 million, five-year cooperative agreement with the U.S. Department of Energy's National Nuclear Security Administration (NNSA) will fund researchers based at Stanford as they strive to develop predictive simulations of hypersonic flight vehicles.
"Predicting phenomena on a computer using simulation technology doesn't require the humongous expenses of physical flight testing and laboratory testing," says Parviz Moin, the Franklin P. and Caroline M. Johnson Professor in the School of Engineering and the project's director. "But hypersonic flight systems cannot be predicted well with today's state-of-the-art simulation capabilities."
Hypersonic flight refers to flight at speeds at least five times faster than the speed of sound, or roughly 3,400 miles an hour at 30,000 feet. Until the last decade, sustained hypersonic flight had only been achieved with rockets, rather than "air-breathing" jet engines. Rockets need to carry their own oxygen source, but jets make use of the oxygen in the atmosphere, making them more practical for transportation.
An example of the kind of plane the scientists will simulate is NASA's X-43, an experimental craft that has flown at Mach 9.6 (or 9.6 times the speed of sound).
Predictive science, the primary focus of the project, is the application of verified and validated computational simulations—an extra assurance of accuracy—to predict the behavior of complex systems where routine experiments are not feasible.
The challenge for the researchers will be to develop algorithms that can model the unique physical phenomena involved in hypersonic travel. Turbulence, aerodynamics, combustion, thermal loads and shockwaves are all different at Mach 10 than in conventional airplanes, or even at the lower end of supersonic flight.
The research effort, announced March 7 as part of the Predictive Science Academic Alliance Program (PSAAP) administered by the National Nuclear Security Administration Office of Advanced Simulation and Computing (NNSA-ASC), involves 16 faculty members from Stanford's departments of Mechanical Engineering, Computer Science, Chemical Engineering, Aeronautics and Astronautics, and Mathematics. Stanford researchers will collaborate with colleagues at the University of Michigan and the State University of New York at Stony Brook on aspects of the work.
Stanford and other PSAAP centers will focus on unclassified applications of interest to the NNSA and its three national laboratories: Lawrence Livermore, Los Alamos and Sandia. The PSAAP centers will develop not only the science and engineering models and software for their large-scale simulations but also methods associated with the emerging disciplines of verification and validation, and uncertainty quantification. The goal of that research is to enable scientists to make precise statements about the degree of confidence they have in their simulation-based predictions.
To perform the simulations, the researchers will have access to some of the NNSA's fastest supercomputers. One such machine, named "Red Storm," has a theoretical peak performance of 124 trillion operations per second and ranks sixth in the world in supercomputer performance, according to the website Top500.org.
The new PSAAP grant follows on a previous NNSA-ASC-funded program at Stanford. In that 10-year, $45 million program, Moin and colleagues developed a comprehensive, integrated simulation of a conventional jet engine.
In that case, as in the new research, Moin said that defining the research with an overarching problem in mind, rather than starting from a set of smaller challenges in particular specialties, has been effective in promoting interdisciplinary research.
"This has been a great model," he says. "How do you get people coming from autonomous departments to come and work together? Having an overarching problem is a great way of doing that."
David Orenstein is the communications and public relations manager at the Stanford School of Engineering.