Search Results

The emerging Variable Valve Timing (VVT) technology complicates the estimation of air flow rate because both intake and exhaust valve timings significantly affect engine's gas exchange and air flow rate. In this paper, we propose to use Artificial Neural Networks (ANN) to model the air flow rate through a 2.4 liter VVT engine with independent intake and exhaust camshaft phasers. The procedure for selecting the network architecture and size is combined with the appropriate training methodology to maximize accuracy and prevent overfitting. After completing the ANN training based on a large set of dynamometer test data, the multi-layer feedforward network demonstrates the ability to represent air flow rate accurately over a wide range of operating conditions. The ANN model is implemented in a vehicle with the same 2.4 L engine using a Rapid Prototype Controller.

A comprehensive quasi-dimensional computer simulation of the spark-ignition (SI) engine was used to explore part-load, fuel economy benefits of the Variable Stroke Engine (VSE) compared to the conventional throttled engine. First it was shown that varying stroke can replace conventional throttling to control engine load, without changing the engine characteristics. Subsequently, the effects of varying stroke on turbulence, burn rate, heat transfer, and pumping and friction losses were revealed. Finally these relationships were used to explain the behavior of the VSE as stroke is reduced. Under part load operation, it was shown that the VSE concept can improve brake specific fuel consumption by 18% to 21% for speeds ranging from 1500 to 3000 rpm. Further, at part load, NOx was reduced by up to 33%. Overall, this study provides insight into changes in processes within and outside the combustion chamber that cause the benefits and limitations of the VSE concept.

Selective Catalytic Reduction (SCR) catalysts used in Lean NOx Trap (LNT) - SCR exhaust aftertreatment systems typically encounter alternating oxidizing and reducing environments. Reducing conditions occur when diesel fuel is injected upstream of a reformer catalyst, generating high concentrations of hydrogen (H₂), carbon monoxide (CO), and hydrocarbons to deNOx the LNT. In this study, the functionality of an iron (Fe) zeolite SCR catalyst is explored with a bench top reactor during steady-state and cyclic transient SCR operation. Experiments to characterize the effect of an LNT deNOx event on SCR operation show that adding H₂ or CO only slightly changes SCR behavior with the primary contribution being an enhancement of nitrogen dioxide (NO₂) decomposition into nitric oxide (NO). Exposure of the catalyst to C₃H₆ (a surrogate for an actual exhaust HC mixture) leads to a significant decrease in NOx reduction capabilities of the catalyst.

An experimental study of the effects of thin ceramic thermal barrier coatings on the performance of a spark-ignited gasoline engine was conducted. A modified 2.5 liter GM engine with ceramic-coated pistons, liners, head, valves and ports was used. Experimental results obtained from the ceramic engine were compared with baseline metal engine data. It was shown that at low-speed part-load conditions encountered in typical driving cycles the ceramic engine could achieve up to 18% higher brake power and up to 10% lower specific fuel consumption. At wide open throttle conditions, the two engines exhibited similar characteristics, except at high speeds where the metal engine showed better performance at the expense of inferior fuel economy. The ceramic coating did not produce any observable knock in the engine and showed no significant wear at the conclusion of the testing phase.

The use of Compressed Natural Gas (CNG) has demonstrated the potential to decrease Particulate Matter (PM) and nitrogen oxide (NOx) emissions simultaneously when used in a dual-fuel application with diesel fuel functioning as the ignition source. However, some authors do find that NOx emissions can increase. One postulation is that the conflicting results in the literature may be due to the difference in composition of natural gas around the world. Therefore, in order to investigate if CNG composition influences combustion performance and emissions, four unique mixtures of CNG were tested (i.e., 87% to 96% methane) while minimizing the combined difference of the density, heating value, and constant pressure specific heat of each mixture. This was accomplished at moderate energy substitution ratios (up to 40%) in a single cylinder engine operating at various loads.

Particulate filters (PF) are a highly effective after-treatment device that reduces particulate matter emissions, a rising environmental concern in the automotive industry. However, accumulation of solid particles during the PF filtration process increases engine backpressure considerably, which can have a negative impact on engine efficiency, acoustics, and gaseous emissions. In this area, an accurate pressure drop model helps to better understand the effect of accumulated solid particles in the PF on engine backpressure, aiding in design and regeneration considerations without physical testing. These effects are further improved on board the vehicle using a single equation pressure drop model with a relatively low computational cost. This article presents a thorough history of PF pressure drop models and their advancements.

Variable Compression Ratio (VCR) technology has long been recognized as a method of improving Spark Ignition (SI) engine fuel economy. The Pressure Reactive Piston (PRP) assembly features a two-piece piston, with a piston crown and separate piston skirt which enclose a spring set between them. The unique feature is that the upper piston reacts to the cylinder pressure, accommodating rapid engine load changes passively. This mechanism effectively limits the peak pressures at high loads without an additional control device, while allowing the engine to operate at high compression ratio during low load conditions. Dynamometer engine testing showed that Brake Specific Fuel Consumption (BSFC) improvement of the PRP over the conventional piston ranged from 8 to 18 % up to 70% load. Knock free full load operation was also achieved. The PRP equipped engine combustion is characterized by reverse motion of the piston crown near top dead center and higher thermal efficiency.

Recently, ozone addition has come under scrutiny as a means of controlling ignition timing for Low Temperature (LTC) combustion, which defeats the NOx-PM tradeoff using a highly dilute, homogeneous mixture. This is because ozone decomposes into atomic oxygen and hydroxyl radicals that influence the early phases of the ignition delay process. In order to understand ozone's influence on combustion better, this work analyzes the effects of ozone-assisted combustion for a single-cylinder, direct-injection Compression Ignition engine via a mechanical pump-line-nozzle fuel system and an electronically controlled common-rail fuel injection system. Experimental outcomes indicate a relatively small influence of ozone for the mechanical injection system with a comparably decreased effect for the common rail system.

Direct injection of natural gas under high pressure conditions has emerged as a promising option for improving engine fuel economy and emissions. However, since the gaseous injection technology is new, limited experience exists as to the optimum configuration of the injection system and associated combustion chamber design. The present study uses KIVA-3 based, multidimensional modeling to improve the understanding and assist the optimization of the gaseous injection process. Compared to standard k-ε models, a Renormalization Group Theory (RNG) based k-ε model [1] has been found to be in better agreement with experiments in predicting gaseous penetration histories for both free and confined jet configurations. Hence, this validated RNG model is adopted here to perform computations in realistic engine geometries.

A detailed chemical kinetic mechanism, consisting of 22 species and 104 elementary reactions, has been used in conjunction with the multi-dimensional reactive flow code KIVA-3 to study autoignition of natural gas injected under compression ignition conditions. Calculations for three different blends of natural gas are performed on a three-dimensional computational grid by modeling both the injection and ignition processes. Ignition delay predictions at pressures and temperatures typical of top-dead-center conditions in compression ignition engines compare well with the measurements of Naber et al. [1] in a combustion bomb. Two different criteria, based on pressure rise and mass of fuel burned, are used to detect the onset of ignition. Parametric studies are conducted to show the effect of additives like ethane and hydrogen peroxide in increasing the fuel consumption rate.

This work reports the development and application of multi-dimensional ignition, combustion and emissions models that account for detailed chemistry and mixing effects in a direct injection engine simulation. A detailed chemical reaction mechanism, consisting of 24 species and 104 reactions, is used for increased accuracy of emissions predictions. Turbulent combustion is represented using a modified Eddy Dissipation Concept (EDC) model to account for mixing effects. The soot model includes all aspects of soot formation and destruction. Particle transport equations are used to realistically track transport of the soot particles formed. All computational sub-models developed in this work have been implemented in a modified version of the KIVA-3V code. In order to illustrate the behavior of the new models, soot and NO emissions have been predicted at different operating conditions by varying injection timing, exhaust gas recirculation (EGR) and injection pressure.

The heat release and pollutant formation processes in a direct injection natural gas engine are studied by coupling detailed chemistry with a multi-dimensional reactive flow code. A detailed kinetic mechanism consisting of 22 species and 104 elementary reactions is chosen by comparing ignition delay predictions with measurements in a combustion bomb. The ignition model is then coupled with a turbulent combustion model and extended Zeldovich kinetics to simulate heat release and nitric oxide production in a direct injection engine. Parametric studies are conducted to investigate the effect of engine operating conditions which include speed, load, injection timing and level of boost. It is shown that use of detailed chemistry is extremely important to predict the correct ignition delay period as engine operating conditions change. Use of both time and crank angle as the independent variable reveals interesting details of the heat release process as a function of engine speed.

The growth of hydraulic fracking has resulted in a dramatic cost reduction of Compressed Natural Gas (CNG), a low carbon fuel. CNG cannot be used as singular fuel in conventional Compression Ignition (CI) engines because of its high auto-ignition characteristics. However, CNG-assisted diesel combustion represents a means to shift the energy consumption of CI engines away from liquid fossil fuels. Calculation of the rate of heat release is vital for understanding and optimizing this mode of engine operation. A previously constructed three-zone equilibrium heat release model that is calibrated to engine exhaust emission measurements was augmented in order to allow for the addition of CNG in the engine intake. The model was also adapted to permit reuse of unburned CNG gas with other exhaust species via exhaust gas recirculation. This is because experiments demonstrated a potentially significant increase in methane emissions under high CNG flowrates.

The effect of natural gas composition on ignition delay has been investigated numerically by using detailed and reduced chemical kinetic mechanisms. Three different blends of natural gas have been analyzed at pressures and temperatures that are typical of top dead center conditions in compression ignition engines. The predicted ignition delay shows a decrease with temperature in an Arrhenius manner and has a first order dependence on pressure. Similar trends have been observed by Naber et al. [1] in their experimental study of natural gas autoignition in a bomb. It is shown that two kinetic mechanisms (GRI-Mech 1.2 and reduced set DRM22) are best capable of predicting the ignition delay of natural gas under compression ignition conditions. The DRM22 mechanism has been chosen for further studies as t involves lower computational costs compared to the full GRI-Mech 1.2 mechanism.

Adjusting the Residual Gas Fraction (RGF) by means of Variable Valve Actuation (VVA) is a strong candidate for controlling the ignition timing in Homogeneous Charge Compression Ignition (HCCI) engines. However, at high levels of residual gas fraction, insufficient mixing can lead to the presence of considerable temperature and composition variations. This paper extends previous modeling efforts to include the effect of RGF distribution on the onset of ignition and the rate of combustion using a multi-dimensional fluid mechanics code (KIVA-3V) sequentially with a multi-zone code with detailed chemical kinetics. KIVA-3V is used to simulate the gas exchange processes, while the multi-zone code computes the combustion event. It is shown that under certain conditions the effect of composition stratification is significant and cannot be captured by a single-zone model or a multi-zone model using only temperature zones.

A combined experimental and analytical approach was followed in this work to study stress distributions and causes of failure in diesel cylinder heads under steady-state and transient operation. Experimental studies were conducted first to measure temperatures, heat fluxes and stresses under a series of steady-state operating conditions. Furthermore, by placing high temperature strain gages within the thermal penetration depth of the cylinder head, the effect of thermal shock loading under rapid transients was studied. A comparison of our steady-state and transient measurements suggests that the steady-state temperature gradients and the level of temperatures are the primary causes of thermal fatigue in cast-iron cylinder heads. Subsequently, a finite element analysis was conducted to predict the detailed steady-state temperature and stress distributions within the cylinder head. A comparison of the predicted steady-state temperatures and stresses compared well with our measurements.

Software packages including Argonne National Lab's Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) model and the EPA's Motor Vehicle Emissions Simulation (MOVES) model are useful in analyzing the emission profiles of light-duty vehicles. In particular, GREET performs a detailed life cycle fuel energy and emissions analysis, while MOVES focuses on energy and emissions during vehicle operation. This study uses MOVES2010b in the creation of emission trends in order to predict future emissions regulations and the subsequent aftertreatment device areas of improvement required to meet these standards. A second objective was to create four time sheet tables in order to update the base vehicle operation emission profiles used in GREET. The simulation results depict the balance between nitrous oxide and hydrocarbon emissions, the dependency upon climate effects, and areas for potential improvement given recent engine design trends.

Lean Premixed Compression Ignition (PCI) low-temperature combustion promises to simultaneously reduce NOx and PM emissions, while suffering a moderate penalty in fuel consumption. Similarly, opportunities exist to develop rich PCI combustion strategies which can provide the necessary exhaust constituents for aggressive regeneration of a Lean NOx Trap (LNT). The current work highlights the development of lean and rich PCI combustion strategies. It is shown that the lean PCI combustion strategy successfully operates with low NOx and PM, at the expense of a 5% increase in fuel consumption over conventional diesel operation. The rich PCI combustion strategy similarly operates with low NOx and PM, and produces enough CO (up to 5% by volume in exhaust) for aggressive regeneration of an LNT.

Biodiesel is a potential alternative to Ultra Low Sulfur Diesel (ULSD); however, it often suffers from increased fuel consumption in comparison to ULSD when injection timings and/or pressures are similar. To decrease fuel consumption, increasing biodiesel injection pressure has been found to mitigate the issues associated with its relatively high viscosity and lower energy content. When doing so, the literature indicates decreased emissions, albeit with potentially greater nitrogen oxide (NOx) emissions in contrast to ULSD. In order to better understand the trade-off between fuel consumption and NOx emissions, this study explores the influence of fuel injection pressure on ULSD, Waste Cooking Oil (WCO) biodiesel, and their blends in a single-cylinder compression ignition (CI) engine. In particular, fuel injection pressures and timings for WCO biodiesel and blended fuels are adjusted to attempt to mimic the in-cylinder pressure profile of operation using ULSD.

A program to explore the effects of natural and additive-derived cetane on various aspects of diesel performance and combustion has been carried out. Procedures have been developed to measure diesel engine fuel consumption and power to a high degree of precision. These methods have been used to measure fuel consumption and power in three heavy-duty direct-injection diesel engines. The fuel matrix consisted of three commercial fuels of cetane number (CN) of 40-42, the same fuels raised to CN 48-50 with a cetane improver additive, and three commercial fuels of base CN 47-50. The engines came from three different U.S. manufacturers and were of three different model years and emissions configurations. Both fuel economy and power were found to be significantly higher for the cetane-improved fuels than for the naturally high cetane fuels. These performance advantages derive mainly from the higher volumetric heat content inherent to the cetane-improved fuels.