Multidimensional Modeling of Fuel Composition Effects on Combustion and Cold-Starting in Diesel Engines 952425

A computer model developed for describing multicomponent fuel vaporization, and ignition in diesel engines has been applied in this study to understand cold-starting and the parameters that are of significant influence on this phenomena. This research utilizes recent improvements in spray vaporization and combustion models that have been implemented in the KIVA-II CFD code. Typical engine fuels are blends of various fuels species, i.e., multicomponent. Thus, the original single component fuel vaporization model in KIVA-II was replaced by a multicomponent fuel vaporization model (based on the model suggested by Jin and Borman). The modelhas been extended to model diesel sprays under typical diesel conditions, including the effect of fuel cetane number variation. Necessary modifications were carried out in the atomization and collision sub-models. The ignition model was also modified to account for fuel composition effects by modifying the Shell ignition model. The improved model was applied to simulate diesel engine cold-starting. The effect of fuel residual from previous cycles was studied and was found to be important. Other injection parameters, such as injection timing and duration were also studied. Another factor that was investigated was engine geometry and how it can be modified to improve on cold-starting in diesel engines. Cold-starting was found to be enhanced by the presence of a small fuel vapor residual and by a shorter injection duration, while engine geometry modifications were found to be helpful in selecting an optimum location on the cylinder head for an ignition aid.

THE SIGNIFICANT NUMBER of diesel engines which provide automotive power, together with recent concerns for the environment, resulted in the introduction of more legislation to limit their pollutant emissions. Of particular interest are the Nitrogen Oxides (NOx) and soot emissions. Both of these pollutants depend significantly on both the fuel and the injection system. The trend to lower emissions motivates new research efforts with the objective of improving engine performance. A significant fraction of engine emissions, particularly unburned hydrocarbons, is produced during the cold-starting phase of the engine operation.

Furthermore, starting the diesel engine under cold ambient conditions represents a difficulty in itself. Thus, more research is needed to identify mechanisms that would improve the cold-startability of the engine. Cold-starting is characterized as a situation where the engine does not fire at all for several cycles, or fires for one cycle and skips firing for the several following cycles, as seen by Henein et al. [1]*. There are several other issues of importance in cold-starting of diesel engines as discussed in Gonzalez et al. [2]. Fuel carryover from cycle-to-cycle is important as liquid fuel remains on the piston surface from misfiring and borderline cycles. Also, another factor in cold-starting in diesels is the excessive wear that results from high peak pressures reached after combustion after misfiring. Blowby gases also tend to increase at the slower cranking speed and that further reduces the compression temperature. Finally, the issue of unburned hydrocarbons and white smoke contributes significantly to the engine emissions. The objective of this work is to understand the process of cold-starting and the various parameters that can help improve the cold-startability of the engine using the most updated and improved computational models in KIVA-II.

The controlling processes in diesel engine combustion are very complicated. To have a better understanding of these processes, they can be broken down to several distinct sub-processes such as injection, atomization, vaporization, ignition, combustion, etc. Once each process is investigated, a comprehensive understanding of diesel combustion can be achieved.

For combustion to start in a fuel spray, the droplets must vaporize partially. Thus, the ignition delay (the delay between the start of injection and the start of combustion), as well as the rate of combustion are strong functions of the droplet vaporization rate [3]. Consequently, the overall engine combustion process and as a result, its efficiency, emissions and cold-starting performance are directly affected by the vaporization process as shown in Ayoub and Reitz [4]. Predicting droplet vaporization accurately would lead to better prediction of the overall combustion process, including both the emissions and starting.

Recently a detailed multidimensional code for modeling reactive flows, such as those encountered in internal combustion engines, has been developed. This code, KIVA-II, includes a model for spray dynamics, a statistical representation that accounts for a spectrum of droplet sizes and the effects of evaporation on sprays, as well as droplet breakup, collision, and coalescence [5]. Spray computations show the sensitivity of the results to the droplet life-time that is, in turn, dependent on the vaporization process. The vaporization models have been extensively discussed by Ayoub and Reitz [4] and results were presented for single droplets with comparison to experimental data, for sprays and several engine cases. The present study utilizes the model improvements that were presented by Ayoub and Reitz [4] in addition to improvements to the ignition model. The goal of this study is to apply the improved models that will be discussed in the next section, to study the sensitivity of diesel combustion to fuel composition as well as investigating factors that are important in cold-starting of diesel engines.