Published in

Elsevier, Proceedings of the Combustion Institute, 1(36), p. 383-391, 2017

DOI: 10.1016/j.proci.2016.08.064

Links

Tools

Export citation

Search in Google Scholar

CFD simulations of Rapid Compression Machines using detailed chemistry: Impact of multi-dimensional effects on the auto-ignition of the iso-octane

This paper was not found in any repository, but could be made available legally by the author.
This paper was not found in any repository, but could be made available legally by the author.

Full text: Unavailable

Green circle
Preprint: archiving allowed
Orange circle
Postprint: archiving restricted
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

voir abstract ; In Rapid Compression Machines (RCM), several phenomena can induce inhomogeneities inside the reaction chamber. The benefits of using a creviced piston have been largely demonstrated through good agreement with the widely used adiabatic core assumption. Still, temperature inhomogeneities due to wall heat transfer cannot be avoided. These induce spatial variations in terms of chemical composition, potentially affecting the auto-ignition process. Mass transfer to the crevices during two-stage ignition is also a phenomenon that can influence the ignition process. In this study, we quantify the impact of multi-dimensional effects on the auto-ignition of the iso-octane by comparing 0-D and RANS simulations of the Argonne RCM. A detailed kinetic mechanism is employed, which makes this study the first to couple an accurate description of both the physical and chemical phenomena in an RCM context for such a complex fuel. It is found that the influence of the inhomogeneities on the ignition delay is globally marginal except for the lowest temperature condition explored where the diffusive transport of intermediate species and radicals plays a key role. The effect of mass transfer to the crevices does not affect significantly the auto-ignition delay under the test conditions. The sensitivity of the results to the turbulence level is also assessed and the results indicate that turbulence may only exert a minor influence on the auto-ignition delay. Comparison with the experimental data is good, and RANS simulation results are similar to those of the 0-D simulations.