Christopher F. Edwards

Dynamic Modeling of Residual-Affected Homogeneous Charge Compression Ignition Engines with Variable Valve Actuation

One practical method for achieving homogeneous charge compression ignition (HCCI) in internal combustion engines is to modulate the valves to trap or reinduct exhaust gases, increasing the energy of the charge, and enabling autoignition. Controlling combustion phasing with valve modulation can be challenging, however, since any controller must operate through the chemical kinetics of HCCI and account for the cycle-to-cycle dynamics arising from the retained exhaust gas. This paper presents a simple model of the overall HCCI process that captures these fundamental aspects. The model uses an integrated Arrhenius rate expression to capture the importance of species concentrations and temperature on the ignition process and predict the start of combustion. The cycle-to-cycle dynamics, in turn, develop through mass exchange between a control volume representing the cylinder and a control mass modeling the exhaust manifold. Despite its simplicity, the model predicts combustion phasing, pressure evolution and work output for propane combustion experiments at levels of fidelity comparable to more complex representations. Transient responses to valve timing changes are also captured and, with minor modification, the model can, in principle, be extended to handle a variety of fuels.

Quasi-steady deformation and drag of uncontaminated liquid drops

Abstract Over 3000 fully resolved numerical simulations have been performed of axisymmetric liquid drops in a uniform gaseous stream. By applying a body force, the drops were held at a fixed velocity relative to the gas in order to determine their quasi-steady deformation response. The solutions were obtained using a new two-fluid spectral/hp finite element method which, as determined by validation tests, is accurate to within 1% for the conditions studied. Liquid-to-gas density ratios between 5 and 500, viscosity ratios between 5 and 15, Weber numbers between 0.1 and 50, and Ohnesorge numbers between 10 −4 and 10 were studied, enabling us to better understand the droplet behavior. Three distinct drop shapes (prolate, oblate, and dimpled) were observed, and the conditions that cause the appearance of these shapes were determined. This allowed a correlation to be developed for predicting the drop shape as a function of the dimensionless parameters governing the system. In addition, a simple criterion predicting the onset of a three-dimensional instability associated with a tumbling motion of the drops was determined. In investigating the drag force on the drops, we found that the current correlations for estimating the effect of internal circulation on the drag for spherical drops were inaccurate and therefore proposed a new correlation. Using this and the deformation correlation, we created a drag model for deforming liquid drops. This model predicts the correct trends in all cases and is usually within 5% of the numerical results. Near break-up, the error becomes larger due to the large deformations in drop shape.

Residual-effected homogeneous charge compression ignition at a low compression ratio using exhaust reinduction

Abstract Studies have been conducted to assess the performance of homogeneous charge compression ignition (HCCI) combustion initiated by exhaust reinduction from the previous engine cycle. Reinduction is achieved using a fully flexible electrohydraulic variable-valve actuation system. In this way, HCCI is implemented at low compression ratio without throttling the intake or exhaust, and without preheating the intake charge. By using late exhaust valve closing and late intake valve opening strategies, steady HCCI combustion was achieved over a range of engine conditions. By varying the timing of both valve events, control can be exerted over both work output (load) and combustion phasing. In comparison with throttled spark ignition (SI) operation on the same engine, HCCI achieved 25–55 per cent of the peak SI indicated work, and did so at uniformly higher thermal efficiency. This was accompanied by a two order of magnitude reduction in NO emissions. In fact, single-digit (ppm) NO emissions were realized under many load conditions. In contrast, hydrocarbon emissions proved to be significantly higher in HCCI combustion under almost all conditions. Varying the equivalence ratio showed a wider equivalence ratio tolerance at low loads for HCCI.

Modeling for control of HCCI engines

The goal of this work is to accurately predict the phasing of homogeneous charge compression ignition (HCCI) combustion for a single cylinder research engine using variable valve actuation (VVA) at Stanford University. Three simple single-zone models were developed and compared with experiment. The difference between the three modeling approaches centered around the combustion chemistry mechanism used in each case. The first modeling approach, which utilized a temperature threshold to model the onset of the combustion reaction, did not work well. However, an integrated reaction rate threshold accounting for both the temperature and concentration did correlate well with experiment. Additionally, another model utilizing a simple two-step kinetic mechanism also showed good correlation with experimental combustion phasing.

Thermodynamic requirements for maximum internal combustion engine cycle efficiency. Part 1: Optimal combustion strategy

Abstract This is the first of a two-part study that examines, from the exergy management standpoint, the fundamental thermodynamic requirements for maximizing internal combustion (IC) engine cycle efficiency. The optimal cycle is shown to comprise three distinct engine architectural elements — reactant preparation, combustion, and work extraction from the products — each of which can be analysed separately. This study shows, based on dynamical system optimization, that it is the equilibrium thermodynamics (specifically, the constant-internal energy—volume (UV) product state at the end of combustion) and not chemical kinetics (i.e. reactions taking place during combustion) that ultimately dictates the amount of exergy destroyed due to combustion. The strategy for minimizing this destruction term reduces to carrying out reactions at the highest possible internal energy state — following what may be called the ‘extreme state’ principle — so as to minimize the corresponding constant-UV entropy change from reactants to equilibrium products. The extreme state principle remains unaltered when system inhomogeneity (from fuel vaporization and mixing with air) and heat loss are accounted for. Based on this optimal combustion strategy, the companion paper examines the remaining elements of the engine cycle (reactant preparation and work extraction) so as to improve overall cycle efficiency.

Structure of a swirl-stabilized spray flame by imaging, laser doppler velocimetry, and phase doppler anemometry

Data are presented which describe the mean structure of a steady, swirl-stabilized, kerosene spray flame in the near-injector region of a research furnace. The data presented include ensemble-averaged results of schlieren, luminosity, and extinction imaging, measurement of the gas phase velocity field by laser Doppler velocimetry, and characterization of the condensed phase velocity by phase Doppler anemometry. The results of these studies define six key regions in the flame: the dense spray region; the rich, two-phase, fuel jet; the main air jet; the internal product recirculation zone; the external product recirculation zone; and the gaseous diffusion flame zone. The first five of these regions form a conical mixing layer which prepares the air and fuel for combustion. The air and fuel jets comprise the central portion of this mixing layer and are bounded on either side by the hot product gases of the internal and external recirculation zones. Entrainment of these product gases into the air/fuel streams provides the energy required to evaporate the fuel spray and initiate combustion. Intermittency of the internal recirculation and spray jet flows accounts for unexpected behavior observed in the aerodynamics of the two phases. The data reported herein are part of a database being accumulated on this spray flame for the purpose of detailed comparison with numerical modeling.

Autoignition of Methanol and Ethanol Sprays under Diesel Engine Conditions

Methanol and ethanol are being considered as alternative fuels for diesel engines. One of the key concerns with using alcohol fuels in diesel engines is their poor ignition quality. This work presents the ignition characteristics of methanol and ethanol examined under simulated diesel engine conditions in a constant-volume combustion vessel. The ignition characteristics of isooctane and normal hexadecane (cetane) measured under the same conditions are also included for reference. Results show that to obtain ignition delays and rates-of-pressure-rise suitable for current diesel engine designs, methanol and ethanol require in-cylinder temperatures of about 1100 K at the time of injection. The results also show that the ignition delays of the alcohol fuels are independent of the chamber-pressure and are unaffected by the presence of 10% by volume of water in the fuel.

Thermodynamic requirements for maximum internal combustion engine cycle efficiency. Part 2: Work extraction and reactant preparation strategies

Abstract This is the second of a two-part study that examines, from the exergy management standpoint, the fundamental thermodynamic requirements for maximizing internal combustion (IC) engine cycle efficiency. In Part 1, it is shown that the strategy to minimize exergy destroyed due to combustion reduces to carrying out combustion at the highest possible internal energy state. Based on this optimal strategy, the present paper examines the remaining elements of IC engine architecture — reactant preparation and product expansion (work extraction) — from the standpoint of managing the associated exergy flows to improve overall engine efficiency. When considered on its own, work extraction is maximized when the combustion products expand to the environmental dead state, with zero exergy left in the exhaust. However, this optimality condition is mismatched to post-combustion conditions for most fuel—air systems, and manifests as hot exhaust with high exergy even upon expansion to ambient pressure. Several strategies to alleviate the mismatch, via preparation of the fuel—air mixture before combustion commences, are considered: reactant compression, dilution with exhaust or excess air, and heating or cooling. These strategies entail trade-offs between exergy destruction due to combustion, and exergy transfers in the form of work (compression), matter (dilution), or heat transfer. The consequent effects on optimal IC engine cycle efficiency are systematically analysed and catalogued.

Understanding chemical effects in low-load-limit extension of homogeneous charge compression ignition engines via recompression reaction

Abstract In-cylinder pre-processing (or recompression reaction) of pilot-injected fuel during negative value overlap (NVO) has been investigated as a method to extend the low-load limit of residual-effected homogeneous charge compression ignition (HCCI). In an effort to elucidate the chemical and thermal effects involved, model calculations have been performed on the recompression reaction and ignition delay of the recompression products using a reduced n-heptane mechanism (160 reactions, 1424 reactions) and a zero-dimensional kinetics model. Parametric studies were performed to cover possible operating choices for HCCI and to understand their effects on the recompression reaction and mixture ignitability. From the study it is demonstrated that the extent of recompression reaction is limited by chemical kinetics, not thermodynamics, and that residual oxygen during NVO is a determining species for the extent and speciation of the recompression reaction. The recompression product mixture exhibits an overall shorter ignition delay than those of the base fuel, except under lean conditions when significant oxidation during NVO leaves only a small amount of fuel available for main ignition. The thermal consequence of the recompression reaction is also largely dependent on oxygen: at near-stoichiometric conditions, the recompression reaction is endothermic from fuel pyrolysis, whereas at lean conditions, the exothermic recompression reaction becomes dominant. Therefore, the chemical and thermal consequences of the recompression reaction exhibit competing effects on mixture ignitability, which leads to an optimum oxygen concentration (equivalence ratio) for reducing ignition delay and extending HCCI operability.

Strategies for Achieving Residual-Effected Homogeneous Charge Compression Ignition Using Variable Valve Actuation

Residual-effected HCCI is investigated using a single-cylinder research engine equipped with fully-flexible variable valve actuation. Dilution limits are explored with various valve profiles in order to gain insight into the best way to use exhaust residual to achieve and control HCCI. The tests repeatedly point out the importance of delayed combustion phasing to reduce thermal losses and maximize efficiency. Combustion phasing is not significantly affected by charge in-cylinder residence time, but is strongly influenced by both the level of exhaust residual and by valve strategies that aim to affect homogeneity. Further dilution with air shows little promise for reaching lower loads, but does suggest that operation near the lean limit can maximize efficiency while minimizing NO and CO emissions. Comparison of exhaust reinduction and exhaust retention (negative valve overlap) strategies shows that, at the same load, reinduction strategies have significantly higher efficiency and reduced NO emissions, though they suffer from higher HC emissions. Reinduction strategies also show some ability to recover from a misfire while retention strategies do not. The data show that the best combination of load range, efficiency, and emissions may be achieved using a reinduction strategy with variable intake lift instead of variable valve timing.