Computational modeling to study the effect of fuel pre-treatment on IC engine combustion control

Abstract

Conventional internal combustion (IC) engine combustion strategies such as homogeneous charge spark ignition (HCSI) and stratified charge compression ignition (SCCI) engines have nearly reached their maximum performance and emission reduction capabilities. New low-temperature combustion (LTC) strategies such as homogeneous charge compression ignition (HCCI) and derivitives have the potential to reduce engine-out emissions while maintaining high efficiency; however, combustion phasing challenges must be solved before their widespread use is adopted. The present work studies the potential of two strategies to control combustion phasing of LTC systems: (1) using intra-cycle re-circulated partial oxidation products (RePOx) and (2) internal fuel reformation by residuals during negative valve overlap (NVO). Both systems were studied using chemical kinetic modeling assuming n-heptane as the fuel. A detailed kinetic mechanism was constructed by combining existing n-heptane and nitrogen mechanisms and validated using HCCI experimental data available from the literature. The RePOx strategy was newly conceived as part of this work. The partial oxidation products are created by extracting a portion of the lean charge products during the expansion stroke and mixing these with the fuel in an auxiliary chamber (RePOx prechamber). The equivalence ratio of the recirculated reactants is controlled by varying the amount of mass extracted. The re-circulated partially-oxidized products are then reintroduced into the main chamber and mixed with compressed air to facilitate the main chamber reaction. This process is modeled using a complex reactor network in the CHEMKIN-PRO software package combined with an external program to balance mass and energy for the RePOx system. The study of this concept was performed in two phases. In the first phase, all the fuel was delivered through the RePOx prechamber, while in the second phase, part of the fuel was premixed in the main chamber prior to compression and the balance was delivered through the prechamber. In both phases, the effects of extraction mass, extraction timing, injection timing, pre-chamber volume, and overall equivalence ratio were examined. Varying pre-chamber volume did not show any effect on the performance or combustion phasing under the conditions and assumptions of this study. In the first phase, advancing injection timing by 5o and 10o crank angle (CA) has advanced the combustion phasing by 1.8o and 3.3o CA respectively. With the premixed charge, the combustion in the main engine chamber exhibited low temperature heat release (LTHR) after 30o crank angle (CA) before top dead center (BTDC) compression. This LTHR varied this trend. When the injection was before LTHR (before -30o CA) the trend is similar to first phase. When the injection is after LTHR (-20o CA), the rise in temperature during LTHR advanced the combustion by 7o CA when compared to -30o CA. In both phases when extraction mass is 5% or above, the combustion is advanced with increased extraction amount. When the extraction mass is below 3%, the incomplete alkane oxidation in pre-chamber caused LTHR in the main chamber after injection causing advanced combustion. Late extraction has delayed the combustion in both phases when there is no LTHR. When there is LTHR, the effect of temperature rise due to LTHR dominated the effect of late extraction and there is no variation in combustion phasing. Increasing overall equivalence ratio without premixing from 04 to 0.5 and 0.6 advanced the combustion phasing by 2o and 3o respectively. Under the conditions of the investigation, the RePOx system without premixing was able to operate at lower overall equivalence ratio than pure HCCI. The (NVO) strategy was incorporated into a ‘conventional’ HCCI engine and was also modeled and evaluated using a complex reactor network in CHEMKIN-PRO. In this case, however, actual experimental data was available from the literature to validate the system as modeled. The data showed that start of injection timing during NVO (NVO_SOI) effected the fuel reformation and varied the main combustion phasing. The main combustion phasing is delayed as the NVO_SOI is intitally retarded since the later injection caused less heat release during NVO, which reduced the temperatures after closing the intake valve (IVC). However, once a particular threshold was reached, additional delay in NVO_SOI resulted in advanced main combustion phasing. The model showed that this was because the reduced time for reformation during NVO caused more alkanes from the reformed fuel to be present during compression of the main combustion event. This triggered low temperature heat release (LTHR) during compression, from which the associated temperature rise caused advanced main combustion. While the model showed the same heat release timing trend as the experimental work, the point of reversing the trend due to LTHR occurred with NVO_SOI 10o crank angle earlier than as it occurred in the experimental results. When both RePOx and NVO systems are compared using the same engine displacement, the RePOx system has more than twice the power output than NVO because the full displacement can be used for fresh charge, whereas the volumetric efficiency is significantly impacted by the NVO valve timing. The RePOx system has more controlling parameters than the NVO system to control the combustion phasing and optimizing performance and emissions. The current research work demonstrates that presence of LTHR effectively minimizes the effect of othe parameters on combustion phasing in both RePOx and NVO systems. LTHR can be minimized by reforming the fuel and controlling the concentrations of species such as HO2, alkenes and alkanes. This work shows that both fuel reforming strategies investigated can be effectively used to control the combustion phasing in LTC systems.