A methodology for diesel engine combustion chamber geometry optimization, using multi-dimensional modeling, has been developed and tested. A flexible chamber geometry parameterization and automated KIVA grid generation technique have been developed and successfully demonstrated. A six-parameter chamber geometry description was developed that allows for a variable squish clearance height at top dead center. A five-parameter chamber description was also developed that maintains a constant squish clearance height.
The chamber geometry optimization methodology was implemented into the KIVA-GA code, which performs multi-dimensional model-based optimizations, subject to design constraints, within the framework of a genetic algorithm. The chamber geometry parameterization and automated grid generation technique was applied to a high-speed direct-injection diesel engine to optimize the combustion chamber geometry at a part-load medium-speed operating condition and a full-load high-speed operating condition. The part-load medium-speed optimization resulted in a novel piston design that featured a unique geometry, which acted to split the fuel distribution between the squish and bowl regions.
The model-based optimizations were followed by an experimental parameric study, which used the part-load medium-speed model-based optimum piston design. The experiments verified that the new piston design had the ability to achieve very low emissions levels, comparable to that of the original factory piston design. Advanced models were then tested in an attempt to improve the agreement of the model predictions with the experimental data.