Search Results

The non-conventional diesel nozzles have attracted more and more attention for their ability to promote jet breakup. In the present study, the internal nozzle flow and jet breakup relying on the convergent-divergent nozzle are investigated by combining the cavitation model and LES model with Multi-Fluid-Quasi-VOF model based on the OpenFOAM code. This is a novel method for which the interphase forces caused by the relative velocity of gas and liquid can be taken into account while sharpening the gas-liquid interface, which is able to accurately present the evolution processes of cavitation and jet breakup. Primarily, the numerical model was verified by the mass flow rate, spray momentum flux, discharge coefficient and effective jet velocity of the prototype Spray D nozzle from the literature.

Flash boiling spray is effective in improving the atomization and evaporation characteristics for gasoline direct injection engines. However, for a multi-hole injector the morphology structure of spray has an obvious change with the fuel temperature increasing or the ambient pressure decreasing, which influences the process of mixture formation and flame propagation. Specially, the spray collapses with both long penetration and a narrow spray angle above certain high superheat degree, which deteriorates air/fuel mixing and hence increases emissions. It is not desired for engine applications while the mechanism of spray structure transformation for multi-hole injector still remains unclear. In the present study, a systematic flash boiling spray model for multi-hole injector is built to investigate the flash boiling spray of multi-hole injector.

To model the spray atomization for diesel HCCI engines, three breakup models including the Taylor Analogy Breakup (TAB), Cascade Atomization and Drop Breakup (CAB) and Kelvin-Helmholtz Rayleigh-Taylor (KH-RT) were evaluated. Based on the experimental results from constant volume, the prediction accuracy of three breakup models was assessed in terms of spray penetration, droplet diameter, droplet velocity and vapor distribution. The results indicate that the mean droplet diameters are significantly underestimated by the TAB model, and the CAB model shows the best performance in the droplet diameter and velocity distributions, but predicts delayed vapor distribution. The KH-RT model shows good predictions in all aspects. By using the KH-RT model, the influence of different injection strategies, including injection timing, spray angle, spray pressure, nozzle hole diameter and split injection, on the mixture preparation process for diesel HCCI engines were investigated.

In situ adaptive tabulation (ISAT) has been implemented into HCCI multidimensional modeling with detailed chemical kinetics, and the performance of ISAT was discussed. The results indicate that ISAT can reduce the computational time remarkably, and the global error can be efficiently controlled. The ISAT without growth and a reversal traverse were tested to ISAT, but they didn't influence the performance of ISAT greatly. Taking account of the character issues of chemical reactions during HCCI combustion process, an enhanced approach, the partial ISAT (PaISAT), was presented, which can significantly improve the accuracy and speed-up factor. The memory occupancy needed by ISAT was reduced based on the dynamic trimming technique.

Dimethyl ether (DME) attracts increasing attentions in recent years, because it can reduce the carbon monoxide (CO), unburned hydrocarbon (HC), and soot emissions for engines as the transportation fuel or the fuel additive. In this paper, a reduced DME oxidation mechanism is developed using the decoupling methodology. The rate constants of the fuel-related reactions are optimized using the non-dominated sorting genetic algorithm II (NSGA-II) to reproduce the ignition delay times in shock tubes and major species concentrations in jet-stirred reactors (JSR) over low-to-high temperatures. In NSGA-II, the range of the rate constants was considered to ensure the reliability of the optimized mechanism. Moreover, an improved objective function was proposed to maintain the faithfulness of the optimized mechanism to the original reaction mechanism, and a new method was presented to determine the optimal solution from the Pareto front.

In this work, large eddy simulation (LES) with a K-equation subgrid turbulent kinetic energy model is implemented into the CFD code KIVA3V to study the features of liquid fuel spray and combustion using gradually varying grid in a constant volume chamber. The characteristic time-scale combustion model (CTC) incorporating a turbulent timescale is adopted to predict the combustion process and the SHELL auto-ignition model is used to predict auto-ignition. Combustion is also simulated using Parallel Detailed Chemistry with Lu's n-heptane reduced mechanism (58 species), which has been added into the KIVA3V-LES code. The computational results are compared with Sandia experimental data for non-reacting and reacting cases. As a result, LES can capture the complex structure of the spray and temperature distribution as well as the trend of ignition delay and flame lift-off length variations. Better results are obtained using the Parallel Detailed Chemistry than the CTC model.

Early injection timing is an effective measure of pre-mixture formation for diesel low-temperature combustion. Three algebraic subgrid models (Smagorinsky model, dynamic Smagorinsky model and WALE model) and one-equation kinetic energy turbulent model using modified TAB breakup model (MTAB model) have been implemented into KIVA3V code to make a detailed large eddy simulation of the atomization and evaporation processes of early injection timing in a constant volume chamber and a Ford high-speed direct-injection diesel engine. The results show that the predictive vapor mass fraction and liquid penetration using LES is in good agreement with the experiment results. In combustion chamber, the sub-grid turbulent kinetic energy and viscosity using LES are less than with the RANS models, and following the increasing time, the sub-grid turbulent kinetic energy and viscosity also increase and are concentrated on the spray area.

Focusing on the dual-mode dual-fuel (DMDF) combustion concept, a combined optimization of the piston bowl geometry with the fuel injection strategy was conducted at low, mid, and high loads. By coupling the KIVA-3V code with the enhanced genetic algorithm (GA), a total of 14 parameters including the piston bowl geometric parameters and the injection parameters were optimized with the objective of meeting Euro VI regulations while improving the fuel efficiency. The optimal piston bowl shape coupled with the corresponding injection strategy was summarized and integrated at various loads. Furthermore, the effects of the key geometric parameters were investigated in terms of organizing the in-cylinder flow, influencing the energy distribution, and affecting the emissions. The results indicate that the behavior of the DMDF combustion mode is further enhanced in the aspects of improving the fuel economy and controlling the emissions after the bowl geometry optimization.

The flow performance of intake port significantly affects engine output power, fuel economy, and exhaust emissions in gasoline engines. Thus, optimal intake port geometry is desired in gasoline engines. To optimize the flow performance of intake port, a new optimization method combining genetic algorithm (GA) and artificial neural network (ANN) was proposed. First, an automatic system for generating the geometry of the tangential intake port was constructed to create various port geometries through inputting the 18 pre-defined structural parameters. Then, the effects of four critical structural parameters were investigated through numerical simulation. On the basis of the computational results, an ANN was used to model the flow performance of the intake port, and a genetic algorithm was simultaneously employed to optimize the flow performance by optimizing the four important structural parameters. Finally, the optimization results were verified through numerical simulation.

Although turbulence plays a critical role in engines operated within low temperature combustion (LTC) regime, its interaction with chemistry on auto-ignition at low-ambient-temperature and lean-oxygen conditions remains inadequately understood. Therefore, it is worthwhile taking turbulence-chemistry interaction (TCI) into consideration in LTC engine simulation by employing advanced combustion models. In the present study, large eddy simulation (LES) coupled with linear eddy model (LEM) is performed to simulate the ignition process in n-heptane spray under engine-relevant conditions, known as Spray H. With LES, more details about unsteady spray flame could be captured compared to Reynolds-averaged Navier-Stokes equations (RANS). With LEM approach, both scalar fluctuation and turbulent mixing on sub-grid level are captured, accounting for the TCI. A skeletal mechanism is adopted in this numerical simulation, including 41 species and 124 reactions.

It has been recognized that density, viscosity, surface tension, and volatility of liquid fuel are of great importance on the atomization and vaporization characteristics of biodiesel spray. This paper presents a comprehensive physical property prediction of biodiesel fuel for spray modeling with most recently developed property prediction models. The temperature-dependent properties of a soy methyl ester (SME) biodiesel were well predicted by the updated prediction methods. Then, the physical properties of the SME biodiesel were added into the KIVA-3V fuel library. By using the well predicted fuel properties, the spray behaviors of SME were successfully simulated by the KIVA-3V code under late-cycle post-injection, conventional diesel injection, and early-injection engine-relevant conditions. The simulation results agree reasonably well with the available experimental liquid penetrations under conditions of various ambient densities and temperatures.

Unlike RANS method, LES method needs more time and much more grids to accurately simulate the spray process. In KIVA, spray process was modeled by Lagrangain-drop and Eulerian-fluid method. The coarse grid can cause errors in predicting the droplet-gas relative velocity, so for reducing grid dependency due to the relative velocity effects, an improved spray model based on a gas-jet theory is used in this work and in order to validate the model seven different size grids were used. In this work, the local dense grid was used to reduce the computation cost and obtain accurate results that also were compared with entire dense grid. Another method to improve computation efficiency is the MUSCL (Monotone Upstream-centered Schemes for Conservation Laws) differencing scheme that was implemented into KIVA3V-LES code to calculate the momentum convective term and reduce numerical errors.

Spray behaviors of pure biodiesel and its blend with conventional diesel have been substantially studied in the last decade. However, the studies on the spray behaviors of pure fatty acid methyl esters (FAMEs) are scarce. The primary components of most biodiesel fuels are methyl palmitate (C16:0), methyl stearate (C18:0), methyl oleate (C18:1), methyl linoleate (C18:2) and methyl linolenate (C18:3), and methyl laurate (C12:0) is also the dominant component of some biodiesels. In this study, the spray behaviors of the aforementioned six FAMEs in biodiesel fuels under engine-relevant conditions were numerically studied using the KIVA-3V code. The physical properties needed for spray modeling were predicted with most recently developed property prediction models and added into the fuel library of KIVA-3V. The transient behaviors of liquid penetrations and vaporization characteristics of these FAMEs were numerically studied under various engine-relevant conditions.

Diesel homogeneous charge compression ignition (HCCI) engines with early injection can result in significant spray/wall impingement which seriously affects the fuel efficiency and emissions. In this paper, the spray/wall interaction models which are available in the literatures are reviewed, and the characteristics of modeling including spray impingement regime, splash threshold, mass fraction, size and velocity of the second droplets are summarized. Then three well developed spray/wall interaction models, O'Rourke and Amsden (OA) model, Bai and Gosman (BG) model and Han, Xu and Trigui (HXT) model, are implemented into KIVA-3V code, and validated by the experimental data from recent literatures under the conditions related to diesel HCCI engines. By comparing the spray pattern, droplet mass, size and velocity after the impingement, the thickness of the wall film and vapor distribution with the experimental data, the performance of these three models are evaluated.