Christophe Barro

INFLUENCE OF POST-INJECTION PARAMETERS ON SOOT FORMATION AND OXIDATION IN A COMMON-RAIL-DIESEL ENGINE USING MULTI-COLOR- PYROMETRY

The emission trade-off between soot and NOx is an issue of major concern in automotive diesel applications. Measures need to be taken both on the engine and on the aftertreatment sides in order to optimize the engine emissions while maintaining the highest possible efficiency. It is known that post injections have a potential for exhaust soot reduction without any significant influence in the NOx emissions. However, an accurate and general rule of how to parameterize a post injection such that it provides a maximum reduction of soot emissions does not exist. Moreover, the underlying mechanisms are not understood in detail. The experimental investigation presented here provides insight into the fundamental mechanisms of soot formation and reduction due to post injections under different turbulence and reaction kinetic conditions. In parallel to the measurement of soot elementary carbon in the exhaust (using a Photo Acoustic Soot Sensor), the in-cylinder soot formation and oxidation process have been investigated with an Optical Light Probe (OLP). This sensor provides crank angle resolved information about the in-cylinder soot evolution The experiments confirm conclusions of earlier works that soot reduction due to a post injection is mainly based on two reasons: increased turbulence (from the post injection) during soot oxidation and lower soot formation due to lower amount of fuel in the main combustion at similar load conditions. A third effect of heat addition during the soot oxidation, which was often mentioned in the literature, could not be confirmed. In addition, the experiments show that variations of turbulence (from swirl) and reaction kinetics have a minor influence on the diffusion controlled heat release rate. However, the time phasing of the soot evolution is highly influenced by these variations with only small changes in the peak soot concentration. It is shown that the soot reduction of a post injection depends on the timing. More precisely, the soot reduction capability of a post injection decreases rapidly as soon as its timing is late in the soot oxidation phase. The soot oxidation rate can only be improved by increased turbulence and heat addition from the post injection in a time window before the in-cylinder soot peak occurs. Depending on EGR and swirl level, a maximum dwell time can be defined after which the post injection effect becomes counterproductive with respect to the soot oxidation rate.

Development and validation of a virtual soot sensor: Part 1: steady-state engine operation

The reduction of the emission limits has lead to an increased complexity of the ECU calibration process and to the need for expensive aftertreatment methods in order to fulfil the legislated limits. Integrating feedback of the Particulate Matter (PM) and NO x emissions into the engine management could make fulfilment of legislation easier and reduce the complexity of the necessary calibration process. Since production type PM sensors for raw emission feedback are not available, a virtual soot sensor (VSS) has been developed. The VSS is a mean value soot model which provides predictions for the PM in real-time (cycle resolved). Its inputs are ECU variables and characteristic values of the heat release rate which are obtained on-line from in-cylinder pressure measurement. The structure of the VSS has been derived from optical kL-measurement data, i.e. from representative, crank angle resolved evolutions of the in-cylinder PM (3-color pyrometry). The model is structured into three consecutive phases which represent the in-cylinder PM evolution and is calibrated with measurements of the exhaust PM concentration of a standard engine operating map only. The three phases correspond to an initial phase where formation of PM dominates, a phase when formation and oxidation are roughly in balance, and a phase during which oxidation dominates. For steady state experiments, the VSS shows an excellent correlation with the exhaust gas soot (respectively elementary carbon) concentration that has been measured with a photo-acoustic soot sensor (PASS). In addition a reasonable ratio between soot formation and soot oxidation is reproduced.

Development and validation of a virtual soot sensor: Part 2: Transient engine operation

The reduction of the legislative emission limits has led to an increased complexity of the engine control unit (ECU) calibration. Integrating feedback of the Particulate Matter (PM) and NOx emissions into the engine management could make fulfilment of legislation easier and reduce the complexity of the necessary calibration process. Due to the fact that production type PM sensors for raw emission feedback will not be available, or will be exceedingly expensive in the near future, in part 1 of this work, a virtual soot sensor (VSS) has been developed. Along with the very good steady state behaviour, the VSS is able to predict PM emissions in transient engine operation with a sufficient precision. This has been approved with different changes in torque demand at constant engine speed. Though an overestimation of the soot occurs during the first cycles of the step in torque demand, the behaviour of the engine out soot was well captured and compared with measurements from a photo-acoustic soot sensor (PASS) and characteristic end values of the representative in-cylinder soot trace (measured by multi-colour pyrometry). The performance of the control structure with integrated VSS is demonstrated on the new European driving cycle and an Urban Dynamometer Driving Schedule. Furthermore, a change in set point value demonstrates the opportunity of changing the raw emission strategy on-line. These results offer the opportunity to expand this cylinder-pressure-based VSS approach to other pollutants (primarily NOx) as well.

Optical Investigations of Soot Reduction Mechanisms using Post-Injections in a Cylindrical Constant Volume Chamber (CCVC)

Past research has shown that post injections have the potential to reduce Diesel engine exhaust PM concentration without any significant influence in the NOx emissions. In earlier research it was observed that soot reduction due to a post injection is based on three reasons: increased turbulence (1) and heat (2) from the post injection during soot oxidation and lower soot formation due to smaller main injection for similar load conditions (3). The second effect of heat addition during the soot oxidation is debated in the literature. The experimental investigation presented in the current work provides insight into the underlying mechanisms of soot formation and reduction using post injections under different operating conditions. The experimental data have been obtained using a cylindrical constant volume chamber with high optical access. The soot evolution has been obtained using 2-color-pyrometry. Furthermore, NO and particle mass and size distribution have been captured from the exhaust. In the experiments, the fuel composition, oxygen content and post injection timing has been varied. All the operating conditions with 21% O2 showed decreased exhaust soot mass for the smallest post injection dwell timing, representing the highest degree of interaction between the two soot clouds. The change in temperature of two merged soot clouds was found to be negligible. Thus, the effect of turbulence induction appears of higher importance than the effect of heat addition. The exhaust soot reduction cannot be observed with reduced O2 background, even with higher degree of interaction. The reason is assumed to be by limited local O2 availability.

Virtuelle Sensoren für Dieselmotoren

Moderne Pkw-Dieselmotoren verfügen über eine Vielzahl von Freiheitsgraden, die einen effizienten Betrieb bei gleichzeitig niedrigen Emissionen ermöglicht. Um das Potenzial dieses komplexen Systems permanent zu nutzen, müssen Emissionen künftig optimal vorgesteuert und geregelt werden. An der ETH Zürich wurden im Rahmen eines FVV-Forschungsvorhabens modellbasierte virtuelle Emissionssensoren hergeleitet, die Informationen über die Emissionen des Motors schneller und günstiger zur Verfügung stellen als herkömmliche physische Emissionssensoren. Diese virtuellen Sensoren erlauben sowohl eine modellbasierte Vorsteuerung als auch eine Integration in einen Regelkreis. Am Prüfstand wurde gezeigt, dass die virtuellen Sensoren im transienten Motorbetrieb als Ersatz für physische Sensoren geeignet sind.