Identification of the impact factor on SI engine cycle by cycle variation and its quasi-dimensional model

The mechanism of gasoline cycle-by-cycle variation (CCV) is investigated in this study using a combined experimental and simulation approach. Experimental data analysis reveals that flame kernel variation is not the sole source of CCV. A strong linear correlation exists between the standard deviation of combustion process timing fluctuations (represented by CA05, CA10, CA50, and CA90) and their mean elapsed combustion times from spark timing. This correlation is independent of engine speeds, loads, dilution ratios and engine type. A quasi-dimensional combustion model is employed for single-parameter simulation study to identify the sources of CCV. The results indicate that fluctuations in turbulence velocity and initial kernel size have the dominant impact on combustion CCV. Based on these findings, a new CCV model has been developed and calibrated. The model demonstrates excellent predictive accuracy for the standard deviation of combustion CCV and indicated mean effective pressure (IMEP) CCV. Specifically, the average absolute error for IMEP CCV prediction is as low as 0.423% and 0.419% across the entire operating range of two different engines.