Experimental and Statistical Comparison of Engine Response as a Function of Fuel Chemistry and Properties in CI and HCCI Engines 2012-01-0857

Knowledge of how fuel chemistry and properties affect engine response is necessary for effective engine control. It may also be possible to tailor fuels to specific combustion modes, engine geometries, or for desired outputs to generate lower emissions and/or higher IMEP and efficiency. Fuel chemistry and properties have different effects on engine performance in CI and HCCI combustion. In this study, experiments were performed using a 517cc Hatz single-cylinder diesel engine and the same engine converted to run in HCCI mode, both equipped with advanced combustion analysis equipment. Engine performance results were modeled statistically with respect to fuel properties, operating parameters, and engine type to determine the extent to which fuel characteristics influence engine response, and how the response differs between the two combustion modes.

Experiments were performed using 16 fuels: ULSD, 9 FACE diesel fuels, and 6 P20 blends of unprocessed plant oils. These fuels were selected because they overlapped the installation of the diesel engine following a long series of HCCI experiments and because they represent a wide range of chemical compositions and properties. This set of fuels was run in CI and HCCI modes across a range of operating conditions. Engine response to the fuels and control parameters was modeled using AVL Cameo statistical modeling software utilizing 2^nd order polynomial models with interactions.

The modeling results demonstrate that HCCI engines respond differently to cetane number and fuel volatility than do CI engines. In addition, engine performance is dictated by fuel chemistry and properties in degrees that differ from CI to HCCI modes. The models are validated and apply equally well to FACE fuels with well-defined fuel characteristics and the P20 plant oils with complex non-traditional fuel characteristics, indicating that the fuel variables selected accurately represent all of the fuels. The models can be used to optimize fuel characteristics and engine control for these specific engines and operating modes.