Mechanical and Civil Engineering Seminar
Combustion currently provides 85% of the US energy needs. In fact, the continued demand for abundant combustion fueled energy will persist well into this century. This places enormous pressure to improve the combustion efficiency in engines for transportation and power generation while simultaneously developing more diverse fuel streams, including carbon neutral biofuels. Ultimately, to shorten the design cycle of new fuels optimally tailored to work with novel fuel efficient, clean engines requires fundamental advances in combustion science underlying the development of predictive models for engineering design. These predictive models couple chemistry with turbulent transport under real world conditions. High performance computing at the petascale has enabled direct numerical simulation (DNS) of simple laboratory scale flames at moderate Reynolds numbers with complex chemistry. Exascale computing on the distant horizon will further enable DNS to be performed at even higher Reynolds number, higher pressure, and with greater chemical complexity.
I will present recent results from petascale DNS focused on fundamental understanding of mixed regimes of combustion relevant to practical gas turbine engines and compression ignition internal combustion engines. These engines are governed by partially-premixed or premixed combustion in the presence of autoignition in a composition and/or thermally stratified environment. Combustion in these mixed regimes is characterized by lean premixed, low temperature, and high pressure thermo-chemical environments where strong coupling between chemical kinetics and turbulent mixing exist. These engines typically burn large hydrocarbon fuels such as gasoline, diesel and jet fuels which exhibit strong low temperature chemical behavior. The low temperature chemistry is modulated by turbulent mixing and, in turn, affects high temperature turbulent flame propagation and structure. Moreover, under intense turbulence environments found in engines mixing with hot products of combustion is often utilized to stabilize a lean flame, for example, through pilot flames, or hot gas recirculation behind a bluff body or in a cavity. This further promotes mixed regimes of combustion through stratification in enthalpy and composition from the product stream.
Finally, with a vision towards exascale combustion simulation I will present recent computer science research highlights from ExaCT, an interdisciplinary exascale combustion co-design center, focused on algorithmic development, programming environment and runtimes, in situ data analytics and uncertainty quantification, and architectural modeling and simulation of combustion application performance.