Stochastic prediction of cycle-by-cycle cylinder pressure fluctuations in a spark ignition engine

Abstract The generality of recent stochastic models for cylinder pressure development is exploited, both to recreate or simulate cyclic datasets with similar statistical properties to real engine data, and to derive many of the more traditional statistics (e.g. standard deviations of maximum pressure, mean effective pressure, burn angle, etc.) directly from the parameters of the more general model form. The same approach also provides insight into the phasing of the cyclic process, since a number of statistics (such as the standard deviation of pressure or burn rate variations) can be computed directly as a function of crank angle. It is also straightforward to evaluate the contribution to such quantities from each of the physically meaningful model parameters, and thereby obtain some insight into the mechanisms involved.