Darius Mehta

Investigation of an In-cylinder Ion Sensing Assisted HCCI Control Strategy

Recent research activities have greatly expanded the understanding of HCCI, its controlling mechanisms, and operation strategies. However, substantially more work is required before HCCI engines will be ready for production. This includes development of a methodology for feedback and closed-loop control of the fuel and air systems to realize HCCI combustion over the speed load range in a production vehicle. In this paper, we use in-cylinder ion sensing to extract the timing of start of combustion and monitor other combustion information such as knocking as feedback signals for closed loop control of HCCI engines. The ion sensor we use is modified from the existing glow plug. This method will minimize the cost relative to an in-cylinder pressure sensor and signal conditioning circuitry while providing equivalent combustion information for the ECU to control the engine. The parameters that affect the ion signal are identified and relationship of ion signal to start of combustion is established, which can be used for HCCI engine control. Ion sensing not only can detect start of fast reaction, but under certain conditions, can also detect the start of cool flame reaction. This correlation could be used to control engine parameters such as charge composition, temperature, and boost of the HCCI engine. However, there is a limitation of when ion sensing is effective and when it is not. At light load, the ion signal strength is too low to be detected. When HCCI combustion is stable and smooth, the ionization signal is stable and the distinction between different combustion phases can be easily identified on the raw ion signal. When HCCI combustion is unstable and violent, the ionization reflects the combustion characteristics as ringing on the ion signal.

Measurement of Laminar Burning Velocity of Multi-Component Fuel Blends for Use in High-Performance SI Engines

A technique was developed for measuring the Laminar Burning Velocity (LBV) of multi-component fuel blends for use in high-performance spark-ignition engines. This technique involves the use of a centrally-ignited spherical combustion chamber, and a complementary analysis code. The technique was validated by examining several single-component fuels, and the computational procedure was extended to handle multi-component fuels without requiring detailed knowledge of their chemical composition. Experiments performed on an instrumented high-speed engine showed good agreement between the observed heat-release rates of the fuels and their predicted ranking based on the measured LBV parameters.