concept simulation Arne Andersson; Bincheng Jiang Volvo Global - - PowerPoint PPT Presentation

Arne Andersson; Bincheng Jiang

Volvo Global Trucks Technology

Marcus Lundgren; Martin Tunér;

Lund University

Karin Fröjd; Simon Bjerkborn; Lars Seidel; Fabian Mauss

Lund Combustion Engineering – LOGE AB

STAR Global Conference 2013

Combustion model for engine concept simulation

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

Combustion model needed for advanced combustion in engine concept development

Systems approach is needed Integration with GT-Power

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

PPC is challenging to model

A wide span of combustion modes in a load sweep

IMEP 8 bar Like HCCI IMEP 12 bar PPC IMEP 26 bar Diffusion combustion

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

Model: 3D – 1D simulation chain

DARS – GT-Power RESULTS DARS SRM + GT-Power

Engine setup

STAR-CD Mixing time (t,x)

OPTIMIZED PROCESS DESIGN

CHEMISTRY

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

The Stochastic Reactor Model (SRM)

The particle properties (Temperature/enthalpy, species concentration) are statistically described by a Probability Density Function (PDF). The development of the PDF is calculated through particle interaction (mixing) and detailed chemistry. In the zero-dimensional Stochastic Reactor Model, the in-cylinder gas mass is discretized into a set of particles without any spatial resolution. Each individual particle is treated as a well-stirred reactor.

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

The Stochastic Reactor Model (SRM)

Mixing is modeled as a stochastic process: Particles are randomly selected to interact with each other. Both temperature and chemical composition is exchanged. The frequency of mixing events is determined by the turbulent mixing time τ

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

The SRM model makes use of detailed chemistry. Three different mechanisms were evaluated:  33-species Tsurushima mechanism  200-species NICE mechanism  477-species Toluene Reference Fuel (TRF) mechanism The 33-species mechanism was found to yield unrealistic ignition timing, whereas the differences between the 200- species mechanism and the 477-species mechanism were negligible. The 200-species NICE mechanism has been used for all simulations in this presentation.

Chemistry

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

 Vaporized fuel is introduced as new particles in the SRM.  These particles are mixed with the background gas (air and EGR) according to the turbulent mixing time.

Fuel injection

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

Example: Vaporized cold fuel mixing with background gas and igniting

T

Ф

T

Ф lg(Xi) lg(Xi)

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

PPC modeling with the SRM

Problem: While the SRM model presumes statistical homogeneity in the combustion chamber, this assumption is not true for the PPC. At the point of ignition, much fuel is still concentrated in a rich zone Solution: Divide the background gas in the SRM into two distinct zones. Let the injected fuel be introduced into one of the zones. Mix predominantly within each zone to capture the effects of stratification.

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

PPC modeling with the SRM

With the two-zone approach, each property of the gas is described by two superimposed PDF:s. The zone mass evolution as function of time is a user input.

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

 The SRM zone mass evolution is calculated from the mixture fraction PDF predicted by STAR-CD – Mass of fuel rich zone = sum of mass in CFD cells where Z > 0.02

Model setup: Zone size definition

CFD data, IMEP 12 bar Zone size comparison, SRM

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

 The mixing time conditioned at stoichiometry is extracted from the STAR-CD simulation and used in the model.

Model setup: Mixing time and other data

 All other relevant parameters were kept constant between the cases.  No further tuning applied.

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

Case 2 – IMEP 8 bar

Time of ignition well predicted. HCCI type combustion

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

Case 1 – IMEP 12 bar

Time of ignition well predicted. A small diffusion tail is seen – PPC like.

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

Case 3 – IMEP 26 bar

Time of ignition pretty well predicted. Diffusion type combustion

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

3D-1D model chain for efficient simulation of HCCI- PPC- and diffusion type combustion  With the right statistical parameters we can create a 0D combustion model capable of PPC simulation – Calculations in STAR-CD are used to obtain the turbulence information needed to set up the model. – A 2-zone stochastic approach is required to match CFD data – HCCI, PPC and diffusion combustion regimes are captured  The Stochastic approach enables the use of large enough kinetic mechanisms  The stochastic approach enables the combustion model integration in GT-Power: full powertrain simulation possible

Sum up and conclusions

Global Trucks Technology

Arne Andersson, STAR Global Conference 2013

Acknowledgements

This material is based upon work supported by – Department of Energy National Energy Technology Lab under Award Number DE-EE0004232 – Department of Energy National Energy Technology Lab under Award Number DE-FC26-07NT43222

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned

• rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily

constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.