This content is not included in your SAE MOBILUS subscription, or you are not logged in.
Analysis of the Performance of a Turbocharged S.I. Engine under Transient Operating Conditions by Means of Fast Running Models
- Journal Article
- DOI: https://doi.org/10.4271/2013-01-1115
ISSN: 1946-3936, e-ISSN: 1946-3944
Published April 8, 2013 by SAE International in United States
Citation: Millo, F., Di Lorenzo, G., Servetto, E., Capra, A. et al., "Analysis of the Performance of a Turbocharged S.I. Engine under Transient Operating Conditions by Means of Fast Running Models," SAE Int. J. Engines 6(2):968-978, 2013, https://doi.org/10.4271/2013-01-1115.
The aim of this work is the assessment of the predictive capabilities of fast running models, obtained through an appropriate reduction and simplification process from detailed 1D fluid-dynamic models, for a turbocharged s.i. engine under highly transient operating conditions.
Simulations results have been compared with experimental data for different types of models, ranging from fully detailed 1D fluid-dynamic models to map-based models, quantifying the degradation of the model accuracy and the reduction in the computational time for different kinds of driving cycles, from moderately transient such as the NEDC to highly dynamic such as the US06.
Although the map based approach was confirmed to be be a viable means to predict fuel economy over the NEDC cycle, even for a downsized and turbocharged engine, thanks to the low accelerations involved and to the almost negligible transients effects, it showed significant discrepancies (error higher than 5%) with the experimental data when applied to highly dynamic driving cycles such as the US06.
On the other hand, the use of fast running models was demonstrated to be a suitable solution to obtain a satisfactory accuracy in the estimate of the fuel consumption also over highly dynamic driving cycles, with differences between the simulation results and the measured values which were within the repeatability range of the experimental tests. Moreover, the reduced computational effort of the fast running models allowed to reach close to real-time simulation performance.