The work presented in this study was performed with the end goal of making advanced combustion engines a more realistic pathway for clean and efficient engines. Two primary challenges of advanced combustion are studied: combustion instability and high load operation. First, the sources of combustion instability are addressed and methods to minimize instability are identified. Second, strategies to achieve high efficiency clean combustion are explored using optimization techniques coupled with KIVA simulations. Along the way several programs were developed to aid in design space exploration and speed up time from conception to completion.
Programs were written to efficiently create hundreds to thousands of detailed CFD simulations and run them all in parallel. Using these programs, the combustion sensitivity of low temperature combustion engines was investigated. Methodologies to predict combustion instability were developed along with investigations in control strategies to reduce it while still maintaining the emissions and efficiency benefit. This work is the stepping stone to develop techniques to quickly and effectively predict combustion instability in advanced combustion engines.
Further work sought to develop a time efficient and effective Genetic Algorithm optimization program to aid in design space exploration. This genetic algorithm was shown to outperform many standard optimization algorithms previously used in this field. The GA was used to develop a low temperature specific piston bowl design and control strategies for both reactivity controlled compression ignition (RCCI) and gasoline compression ignition (GCI) engines.