338011.pdf (28.2 MB)
Computational fluid dynamic modelling of flow and combustion in spark ignition engines
thesis
posted on 2010-11-24, 11:17 authored by Sudhakar DasThe present work is based on the need for understanding the in-cylinder flow and its
subsequent effects on combustion in a valved-two-stroke spark ignition engine with
fuel injection using Computational Fluid Dynamics (CFD) and experimental
techniques. In this context, the CFD code KIVA-II has been modified to model the
two-stroke engine gas exchange and combustion processes. A 3-D Cartesian grid
generation program for complex engine geometry has been added to the KIVA code
which has been modified to include intake and exhaust flow processes with valves.
New and improved sub models for wall jet interaction, mixing controlled combustion
and one dimensional wave action have also been incorporated. The modified version of
the program has been used to simulate a fuel injected two-stroke spark ignition engine
and parametric studies have been undertaken. The simulated flow, combustion and
exhaust emission characteristics over a wide range of operating conditions show the
expected trends in behaviour observed in actual engines.
In the second phase of this study, the air-assisted-fuel-injection (AAFI) process into a
cylinder has been simulated with a high resolution computational grid. The simulation
results are presented and compared with experimental data obtained using the Schlieren
optical technique. An approximate method based on the conservation of mass,
momentum and energy of the spray jet and using a comparatively coarse grid has been
suggested for simulating the AAFI process. The simulation study predicts a high degree
of atomisation of fuel spray with Sauter mean diameter around 10 μm even with
moderate air and fuel pressures. The penetration and width of spray are simulated within
15% of the experimental values.
In the last phase of this study, the flow and combustion processes have been studied for
a four-stroke spark ignition engine with the AAFI process. The simulation results
obtained using this approximate method have been validated with experimental data
generated for the same engine configuration.
History
School
- Aeronautical, Automotive, Chemical and Materials Engineering
Department
- Aeronautical and Automotive Engineering
Publisher
© Sudhakar DasPublication date
1996Notes
A Doctoral Thesis. Submitted in partial fulfillment of the requirements for the award of Doctor of Philosophy of Loughborough University.EThOS Persistent ID
uk.bl.ethos.338011Language
- en