Conceptual Limitations of the Probability Density Function Method for Modeling Turbulent Premixed Combustion and Closed-Form Description of Chemical Reactions’ Effects

In this paper, we critically analyzed possibilities of probability density function (PDF) methods for the closed-form description of combustion chemical effects in turbulent premixed flames. We came to the conclusion that the concept of a closed-form description of chemical effects in the classical modeling strategy in the PDF method based on the use of reaction-independent mixing models is not applicable to turbulent flames. The reason for this is the strong dependence of mixing on the combustion reactions due to the thin-reaction-zone nature of turbulent combustion confirmed in recent optical studies and direct numerical simulations. In this case, the chemical effect is caused by coupled reaction–diffusion processes that take place in thin zones of instantaneous combustion. We considered possible alternative modeling strategies in the PDF method that would allow the chemical effects to be described in a closed form and came to the conclusion that this is possible only in a hypothetical case where instantaneous combustion occurs in reaction zones identical to the reaction zone of the undisturbed laminar flame. For turbulent combustion in the laminar flamelet regime, we use an inverse modeling strategy where the model PDF directly contains the characteristics of the laminar flame. For turbulent combustion in the distributed preheat zone regime, we offer an original joint direct/inverse modeling strategy. For turbulent combustion in the thickened flamelet regime, we combine the joint direct/inverse and inverse modeling strategies correspondingly for simulation of the thickened flamelet structure and for the determination of the global characteristics of the turbulent flame.

Experimental investigation of combustion and heat transfer in a direct-injection spark ignition engine via instantaneous combustion chamber surface temperature measurements

An experimental study was performed to provide the combustion and in-cylinder heat transfer characteristics resulting from different injection strategies in a direct-injection spark ignition (DISI) engine. Fast-response thermocouples were embedded in the piston top and cylinder head surface to measure the instantaneous combustion chamber surface temperature and heat flux, thus providing critical information about the combustion characteristics and a thorough understanding of the heat transfer process. Two distinctive operating modes, homogeneous and stratified, were considered and their effect on combustion and heat transfer in a DISI engine was investigated. The stratified operating mode yielded significantly higher spatial variations of heat flux than the homogeneous mode. This behaviour is directly caused by the main features of stratified combustion, i.e. vigorous burning of a close-to-stoichiometric mixture near the spark, and a cool, extremely lean mixture at the periphery. The cooling effect of the spray impinging on the piston surface when the fuel is injected late in compression could be detected too. The local phenomena change with varying speed and injection parameters. Comparison between the calculated global heat fluxes and measured local heat fluxes were performed in order to assess the behaviour of classic heat transfer models. Comparisons between the global and local heat fluxes provide additional insight into spatial variations, as well as indications about the suitability of different classic models for investigations of the heat transfer aspect of DISI engines. Special consideration is required when applying classic heat transfer correlations to stratified DISI operation as heat flux values are lower by more than 30 per cent when compared with homogeneous operation of the same engine at the same load.