Computational fluid dynamic analysis of knock onset in diesel dual-fuel engine

• Main Author: Ghiji, Mohammadmahdi
• Format: Thesis
• Language: English
• Published: 2011
• Online Access: http://psasir.upm.edu.my/id/eprint/41801/1/FK%202011%2026R.pdf; http://psasir.upm.edu.my/id/eprint/41801/

Energy alternative for the transport sector includes use of natural gas due its large reserves and minimum modifications to the existing diesel engines. The most common diesel dual fuel (DDF) engines use the natural gas as main fuel to provide the engine output power while a low amount of diesel fuel injected to the combustion chamber as the ignition source of the air-gaseous fuel mixture. The DDF engines suffer from low thermal efficiency during part load operations and knock tendency during high load operations compare to the conventional diesel fuel. Improving the DDF engine performance needs the clear understanding of occurring phenomena such as autoignition, injection of diesel fuel and combustion of dual fuel mixture in combustion chamber. Therefore, besides the experimental study, numerical analyses are vital to provide an understanding into the complex process inside the combustion chamber. Hence, the specific objectives of this study are (i) employing the numerical simulation of DDF engine to investigate the effects of different intake mixture temperatures and mixing ratios on the engine performance and knock intensity by using in-cylinder pressure analysis (ii) comparing the accuracy of the chosen turbulence models, k-e standard and k-e Re-Normalized Group (RNG), due to the experimental data. For examining the effect of increased intake mixture (57°C, 70°C and 90°C) and mixing ratio of natural gas and diesel fuel (83%, 85%, 87% and 90%) on performance and knock intensity of DDF engine, a prediction of in-cylinder pressure during engine cycle (340-400 degree of crank angle) using Computational Fluid Dynamics (CFD) technique code, Fluent, was performed and compared the extracted results to the experimental data of the previous researchers as mentioned in literature review chapter on a single cylinder Ricardo Hydra engine. The initial grid was created in Gambit software. CFD codes written for describing the intake and exhaust valve movement were applied to probe the in-cylinder air motion and direct injection of diesel fuel during the intake and compression strokes. The Moving Dynamic Mesh model was performed for the grid generation due to the moving mesh and boundary to provide a more accurate transient condition. A Lagrangian Particle Based approach described the diesel fuel spray and Wave model utilized to represent both the primary breakup (liquid atomization) and the secondary breakup (drop breakup) processes. The combustion process is modeled with the Eddy Dissipation Model of Magnussen and Hjertager for partially premixed reaction with five global reaction schemes. The K-e standard and k-e RNG turbulence models were used for the turbulent fluid flows inside the combustion chamber and the predicted results were achieved through the sequential process which ensured accuracy of the computations. The results demonstrated that, the k-e RNG turbulence model presented more accurate results for predicting the in-cylinder pressure; also the obtained results verified the reliability engines, would lead us to an engines.