Instantaneous within cycle model based fault estimators for SI engines

The Mean Value Engine Model, commonly used in model based fault estimation of automotive spark ignition gasoline engines describes only the averaged dynamics of the engines, resulting in reduced sensitivity and isolability of faults. In this paper, estimation of faults is done using an instantaneous physics based within cycle model of the gasoline engine. A novel method for estimating certain types of faults is proposed and validated against the standard industrial simulation software, AMESim. The results of fault estimation show the efficacy of the method and also signify the importance of their instantaneously pulsating nature for characterizing the true nature of the fault.