Sandia researchers develop autoignition model designed for efficient, accurate engine simulations
Sandia researchers have developed a model for autoignition in diesel engines that plays an important role in determining engine efficiency and formation of pollutants.
Sandia postdoctoral researcher Layal Hakim, working with mentors and Sandia principal investigators Guilhem Lacaze and Joe Oefelein (all 8351), designed and implemented an optimized chemical model that describes autoignition of a diesel fuel surrogate with quantified uncertainty quantification. This model is a key component in developing simulations that provide an unprecedented level of insight into the effect of high-pressure liquid injection, fuel chemistry, and turbulent mixing on diesel combustion efficiency and emissions characteristics.
The research, published in April in the SAE International Journal of Fuels and Lubricants, is titled “Large Eddy Simulation of Autoignition Transients in a Model Diesel Injector Configuration.”
“Given environmental concerns, there’s a pressing need to develop more efficient and cleaner engines and fuels,” Layal says. “One bottleneck is understanding oxidation of large hydrocarbon fuels over a wide range of operating conditions.”
A surrogate fuel
Because diesel fuels are composed of thousands of chemical species, detailed kinetic models are typically not practical for direct use in simulations. Thus, surrogate fuels composed only of a few components are used to approximate the physical and chemical properties of real fuels.
“We use n-dodecane in our simulation as a surrogate fuel to mimic diesel. But while detailed mechanisms are an active research topic to model and understand the chemical behavior of such surrogates, we still need a more affordable representation of the subtleties of the n-dodecane chemistry when we study the key physics that lies in the combination of mixing and chemistry. This is where our chemical model has proven its usefulness,” she says. “The physical processes in diesel jet injection and ignition are still not fully understood and experiments, while invaluable, can only provide a limited level of detail. Therefore, numerical simulations are good candidates to reveal missing information.”
The group collaborated with Mohammad Khalil, Khachik Sargsyan, and Habib Najm (all 8351) for their expertise in uncertainty quantification.
“Quantifying the impact of uncertainties introduced by such a chemical approximation of the simulation predictions is crucial for providing meaningful data,” Layal says.
Engine optimization
Understanding the fundamental processes that lead to autoignition can help engine design and optimization because the timing and location of ignition have a direct bearing on system efficiency and emissions. The spatial and temporal fidelity of these calculations provide access to full broadband three-dimensional characteristics of injection, ignition, and combustion.
The lack of accurate models representing the physics of injection, vaporization, turbulent mixing, and ignition is a major barrier to the design of new engines. Thus, simulations conducted for this research can provide complementary high-resolution data beyond what can be measured in experiments to better understand diesel jet mixing and ignition. The simulations aim to complement key experiments by providing benchmark data at the same conditions to test and improve engineering approaches used in industry.
Studies such as these contribute to the whole engine community through Sandia’s Engine Combustion Network.
A future goal is to perform joint comparisons that use data generated here to understand the accuracy of models used in engineering codes.
This published research fits into the philosophy of Sandia’s Combustion Research Facility where simulations complement experiments and bring key insights to improve real engines.