IFPEN utilise des cookies tiers afin de mesurer l’audience et le contenu proposé sur ce site web. IFPEN utilise également des cookies tiers pour vous permettre d’interagir avec les réseaux sociaux et de visualiser du contenu multimédia sur le site. Vous pouvez refuser ou accepter l’ensemble de ces cookies. Vous pouvez également exprimer votre choix par finalité via "Paramétrer les cookies".

06.01.2020

5 minutes de lecture

FacebookTwitterLinkedInImprimer

Future constraints on pollutant emissions pushed car manufacturers towards Direct-Injection Spark-Ignition (DI-SI) technologies to improve engine performances and reduce both fuel consumption and emissions. New challenges are then introduced in terms of combustion optimization, due to a more complex phenomenology and a larger number of degrees of freedom, while system models require developments to approach such new engine architectures with the sufficient level of detail.

A combustion model that works toward a more comprehensive approach

In order to achieve the latter objective, a PhD work performed at IFPEN[1] hosted the development and validation of a Zero-Dimensional (0D) model of DI-SI combustion for system simulation. The proposed model focuses on physics of atomization and drop evaporation, fuel/air mixing, flame propagation in heterogeneous charge and mutual interaction between these phenomena, Figure 1.
 

Figure-1-Synoptic-diagram-of-the-modelling-approach
Figure 1: Synoptic diagram of the modelling approach

In particular, it is characterized by the following aspects:
  

  • The liquid phase is discretized in parcels grouping drops of the same size. An empirical atomization model based on injection velocity, fuel characteristics and thermodynamic conditions provides initial diameters. A Lagrangian model including drag-inertia dynamics, heat-up and forced convection describes drop parcel penetration and evaporation.
       
  • Fuel / air mixing is described using a discrete Probability Density Function (PDF) approach, based on constant-mixture-fraction classes interacting with each other and with the drop parcels[2].
       
  • Flame propagation takes into account mixture heterogeneity effects on flame speed and pollutant production is modeled.


A validation method based on experimental results and simulation tests

The approach was validated against experimental results, when available, and 3D CFD*  RANS** simulations[3], Figure 2 and Figure 3.

Figure-2-Comparison-of-the-gaseous-and-liquid-penetrations
Figure 2: Comparison of the gaseous (left) and liquid (right) penetrations 0D model results to experiments (Schlieren for gaseous penetration and Mie Scattering for liquid penetration) and 3D CFD results, in a constant volume vessel (operating conditions are P=1.54bar and T=388K; fuel is isooctane).

 

Figure-3-Comparison-of-fuel-mixture-fraction-distribution
Figure 3: Comparison of fuel mixture fraction distribution 0D-model results to 3D CFD results, in a DI-SI engine at different crank angles instants: a, 270 CAD***  BTDC**** ; b, 180 CAD BTDC; c, 90 CAD BTDC; d, 0 CAD BTDC. Operating conditions of the engine are 6bar of IMEP and 1200rpm of rotational speed; injection starts at 278.6 CAD BTDC.

*      CFD  = Computational Fluid Dynamics
**    RANS  = Reynolds-Averaged Navier–Stokes
***  270 CAD =  Crank Angle Degree 
**** BTDC =   Before Top Dead Center

 

The implemented model already opens up multiple perspectives

The model was implemented in the Simcenter Amesim platform for multi-physical modelling of the Siemens Digital Industries Software and integrated in the CFM Spark Ignition combustion chamber submodel of the IFP-Engine library.

This PhD work opens various perspectives for future works:
  

  • The discrete PDF, only applied so far to the fresh mixture, could be employed to describe the mixing-controlled post-oxidation reactions in the exhaust gas, thus improving the gaseous pollutant emissions prediction capability of the model;
      
  • The parcel-based liquid phase model could be used to predict the mass of liquid fuel colliding with the cylinder walls, information that can be used to develop a liquid film model;
      
  • Predicting mass and drops size of liquid fuel present in the cylinder during combustion is also highly relevant in the context of soot formation modelling.
     

Scientific Contact: Alessio Dulbecco


Publications

[1] F. Pellegrino, System Simulation of Combustion in Direct-Injection Spark-Ignition Engines, CentraleSupelec PhD thesis, 2019.
  
[2] F. Pellegrino, A. Dulbecco, D. Veynante, Development of a Quasi-Dimensional Spray Evaporation and Mixture Formation Model for Direct-Injection Spark-Ignition Engines, SAE Technical paper, 2015
>> DOI: 10.4271/2015-24-2471
   
[3] F. Pellegrino, A. Dulbecco, D. Veynante, Development and validation of a quasi-dimensional spray model for DI-SI engines, Thiesel conference on Thermo-and-Fluid Dynamic Processes in Direct Injection Engines, poster session, 2018.