Search Results

Accurate modeling of the internal flow and spray characteristics in fuel injectors is a critical aspect of direct injection engine design. However, such high-fidelity computational fluid dynamics (CFD) models are often computationally expensive due to the requirement of resolving fine temporal and spatial scales. This paper addresses the computational bottleneck issue by proposing a machine learning-based emulator framework, which learns efficient surrogate models for spatiotemporal flow distributions relevant for static coupling injection maps, namely total void fraction, velocity, and mass, within a design space of interest. Different design points involving variations of needle lift, fuel viscosity, and level of non-condensable gas in the fuel were explored in this study. An interpretable Bayesian learning strategy was employed to understand the effect of the design parameters on the void fraction fields at the exit of the injector orifice.

Fuel supercriticality has recently received significant attention due to the elevated pressures and temperatures that directly-injected (DI) fuel sprays encounter in modern internal combustion (IC) engines. This paper presents a theoretical examination of conventional and alternative DI fuels at conditions relevant to the operation of compression ignition (CI) engines. The focus is to identify the conditions under which the injected liquid fuel can bypass the atomization process and directly transition to a diffusional mixing regime with the chamber gas. Evaluating the microscopic length-scales of the phase boundary associated with the injection of liquid nitrogen into its own vapor, it is found that the conventional threshold based on the interfacial Knudsen number (i.e. Kn = 0.1) does not adequately quantify the direct transition between sub- and supercriticality. Instead, a threshold that is an order of magnitude smaller is more appropriate for this purpose.

End-of-injection transients have recently been shown to be important for combustion and emissions outcomes in diesel engines. The objective of this work is to develop an understanding of the coupling between end-of-injection transients and the propensity for second-stage ignition in mixtures upstream of the lifted diesel flame, or combustion recession. An injection system capable of varying the end-of-injection transient was developed to study single fuel sprays in a newly commissioned optically-accessible spray chamber under a range of ambient conditions. Simultaneous high-speed optical diagnostics, namely schlieren, OH* chemiluminescence, and broadband luminosity, were used to characterize the spatial and temporal development of combustion recession after the end of injection.

Spray processes, such as primary breakup, play an important role for subsequent combustion processes and emissions formation. Accurate modeling of these spray physics is therefore key to ensure faithful representation of both the global and local characteristics of the spray. However, the governing physical mechanisms underlying primary breakup in fuel sprays are still not known. Several theories have been proposed and incorporated into different engineering models for the primary breakup of fuel sprays, with the most widely employed models following an approach based on aerodynamically-induced breakup, or more recently, based on liquid turbulence-induced breakup. However, a complete validation of these breakup models and theories is lacking since no existing measurements have yielded the joint liquid mass and drop size distribution needed to fully define the spray, especially in the near-nozzle region.

Diesel combustion and emissions is largely spray and mixing controlled. Spray and combustion models enable characterization over a range of conditions to understand optimum combustion strategies. The validity of models depends on the inputs, including the rate of injection profile of the injector. One method to measure the rate of injection is to measure the momentum, where the injected fuel spray is directed onto a force transducer which provides measurements of momentum flux. From this the mass flow rate is calculated. In this study, the impact of impingement distance, the distance from injector nozzle exit to the anvil connected to the force transducer, is characterized over a range of 2 - 12 mm. This characterization includes the impact of the distance on the momentum flux signal in both magnitude and shape. At longer impingement distances, it is hypothesized that a peak in momentum could occur due to increasing velocity of fuel injected as the pintle fully opens.

Experimental studies have been undertaken on a single-cylinder HPCR diesel engine with a compression ratio of 15.5:1 to explore the effect of fuel injection strategy on cycle by cycle stability. The influence of the number, separation and quantity of pilot injections on the coefficient of variation of IMEP has been investigated at -20°C, 1000 rev/min, post-start idling conditions. Injection strategy and glow plug temperature trade-off has also been investigated at a range of soak temperatures. Up to four pilot injections have been used. For timing of the main injection near to the optimum, CoVIMEP values of 10% or better can be achieved. Closer spacing of injections improved stability and extended the range of timings to meet target stability. The best combinations of pilot number and pilot quantity varied with total fuel delivered.

U.S. federal vehicle emission standards effective in 2007 require tight control of NOx and hydrocarbon emissions. For light-duty vehicles, the current standard of Tier 2 Bin 5 is about 0.07 g/mi NOx and 0.09 g/mi NMOG (non-methane organic gases) at 120,000 mi. However, the proposed future standard is 0.03 g/mi for NMOG + NOx (~SULEV30) at 150,000 mi. There is a significant improvement needed in catalyst system efficiencies for diesel vehicles to achieve the future standard, mainly during cold start. In this study, a less than 6000 lbs diesel truck equipped with an advanced urea Selective Catalytic Reduction (SCR) system was used to pursue lower tailpipe emissions with an emphasis on vehicle calibration and catalyst package. The calibration was tuned by optimizing exhaust gas recirculation (EGR) fuel injection and cold start strategy to generate desirable engine-out emissions balanced with reasonable temperatures.

Soot formation and distribution inside the cylinder of a light-duty direct injection diesel engine, have been predicted using Kiva-3v CFD software. Pathlines of soot particles traced from specific in-cylinder locations and crank angle instants have been explored using the results for cylinder charge motion predicted by the Kiva-3v code. Pathlines are determined assuming soot particles are massless and follow charge motion. Coagulation and agglomeration have not been taken into account. High rates of soot formation dominate during and just after the injection. Oxidation becomes dominant after the injection has terminated and throughout the power stroke. Computed soot pathlines show that soot particles formed just below the fuel spray axis during the early injection period are more likely to travel to the cylinder wall boundary layer. Soot particles above the fuel spray have lesser tendency to be conveyed to the cylinder wall.

The fuel-ambient mixture in vaporized fuel jets produced by liquid sprays is fundamental to the performance and operation of engines. Unfortunately, experimental difficulties limit the direct measurement of local fuel-ambient mixture, inhibiting quantitative assessment of mixing. On the other hand, measurement of global quantities, such as the jet penetration rate, is relatively straightforward. Simplified models to predict local fuel-ambient mixture have also been developed, based on these global parameters. However, experimental data to validate these models over a range of conditions is needed. In the current work, we perform measurements of jet global quantities such as vapor-phase penetration, liquid-phase penetration, spreading angle, and nozzle flow coefficients over a range of conditions in a high-temperature, high-pressure vessel.