No abstract

Combustion of a charge with spatially and temporally varying equivalence ratio in a spark ignition engine was modelled using the Leeds University Spark Ignition Engine quasi-dimensional thermodynamic code. New sub-models have been integrated into Leeds University Spark Ignition Engine that simulate the effect of burnt gas expansion and turbulent mixing on an initial equivalence ratio distribution. Realistic distribution functions were used to model the radially varying equivalence ratio. The new stratified fuel model was validated against experimental data, showing reasonable agreement for both the pressure trace and percentage heat released. Including the effect of turbulent mixing was found to be important to reproduce the trend in the differences between the stratified and homogeneous simulations.

Reported in the current paper is a study into the cycle efficiency effects of utilising a complex valvetrain mechanism in order to generate variable in-cylinder charge motion and therefore alter the dilution tolerance of a Direct Injection Spark Ignition (DISI) engine. A Jaguar Land Rover Single Cylinder Research Engine (SCRE) was operated at a number of engine speeds and loads with the dilution fraction varied accordingly (excess air (lean), external Exhaust Gas Residuals (EGR) or some combination of both). For each engine speed, load and dilution fraction, the engine was operated with either both intake valves fully open-Dual Valve Actuation (DVA)-or one valve completely closed-Single Valve Actuation (SVA) mode. The engine was operated in DVA and SVA modes with EGR fractions up to 20% with the excess air dilution (Lambda) increased (to approximately 1.8) until combustion stability was duly compromised. At 1500 Revolutions Per Minute (RPM), 3.6bar and 7.9bar Gross Mean effective Pressure (GMEP), the dilution tolerance of the engine was significantly increased for a given combustion stability limit utilising SVA. This resulted in fuel consumption reductions of up to 3.8% and 3.1% respectively for these two engine speed and load conditions as a result of being able to operate the engine with more thermodynamically attractive mixtures when adopting SVA. At 2000RPM, 9.8bar GMEP, the dilution tolerance was only marginally increased which resulted in a fuel consumption reduction of 1.3% when adopting SVA over DVA (for the same reasons outlined above). Increased dilution tolerance in all cases was achieved as a result of significant enhancement in charge motion when adopting SVA. By enhancing the in-cylinder charge motion (confirmed using Computational Fluid Dynamics (CFD)), ignition to 10% Mass Fraction Burned (MFB) and 10-90% MFB durations for equivalent levels of dilution were significantly shorter when adopting SVA. This therefore allowed greater dilution tolerance (and ultimately an increase in the thermal efficiency of the working cycle) when adopting SVA over DVA without detrimental increases in the burn duration metrics that would ordinarily result in misfire and partial burn and a significant detriment to combustion stability. Conversely, for equivalent levels of dilution, there was little, if any difference in fuel consumption between DVA and SVA even though burn duration metrics were significantly shorter when adopting SVA over DVA. In combination with CFD, the polytropic coefficient of compression was calculated to be lower in all cases for SVA compared to DVA for a given level of dilution. This indicated greater heat transfer when adopting SVA over DVA for equivalent trapped mass (confirmed using CFD). As such, this detrimental increased heat transfer (again confirmed with CFD) attributed to the increased in-cylinder activity with SVA offset the favourably faster combustion; thus resulting in little, if any reduction in fuel consumption for equivalent levels of dilution when implementing SVA over DVA. This...

The article compares the accuracy of different exhaust gas recirculation (EGR) correction factor models under engine conditions. The effect of EGR on the laminar burning velocity of a EURO VI E10 specification gasoline (10% Ethanol content by volume) has been back calculated from engine pressure trace data, using the Leeds University Spark Ignition Engine Data Analysis (LUSIEDA) reverse thermodynamic code. The engine pressure data ranges from 5% to 25% EGR (by mass) with the running conditions, such as spark advance and pressure at intake valve closure, changed to maintain a constant engine load of 0.79 MPa gross mean effective pressure (GMEP). Based on the experimental data, a correlation is suggested on how the laminar burning velocity reduces with increasing EGR mass fraction. This correlation, together with existing models, was then implemented into the quasi-dimensional Leeds University Spark Ignition Engine (LUSIE) predictive engine code and resulting predictions are compared against measurements. It was found that the new correlation is in good agreement with experimental data for a diluent range of 5%-25%, providing the best fit for both engine loads investigated, whereas existing models tend to overpredict the reduction of burning velocity due to EGR.