Search Results

Accurate modeling of the characteristics of diesel-engine combustion leads to more efficient design. Accurate modeling in turn depends on correctly capturing spray dynamics, turbulence, and fuel chemistry. This work presents a computational fluid dynamics (CFD) investigation of a well characterized small-bore direct injection diesel engine at Sandia National Laboratories’ Combustion Research Facility. The engine has been studied for two piston-bowls geometries and various injection timings. Simulation of these conditions test the predictive capabilities of our approach to diesel engine modeling using Ansys Forte. An experimental database covering a wide range of operating conditions is provided by the Engine Combustion Network for this engine, which is used to validate our modeling approach. Automatic and solution-adaptive meshing is used, and the recommended settings are discussed.

Flash boiling occurs in sprays when the ambient gas pressure is lower than the saturation pressure of the injected fuel. In the present work, a numerical study was conducted to investigate solid-cone spray behaviors under various flash-boiling conditions. A new spray cone angle correlation that is a function of injection parameters was developed and used for spray initialization at the nozzle exit to capture plume interactions and the global spray shape. The spray-breakup regime control was adjusted to enable catastrophic droplet breakup, characterized by Rayleigh-Taylor (RT) breakup, near the nozzle exit. The model was validated against experimental spray data from five different injectors, including both multi-hole and single-hole injectors, with injection pressure varying from 100 to 200 bar.

This study focuses on Lagrangian spray models that are commonly used in engine CFD simulations. In modeling sprays, droplet collision is one of the physical phenomena that must be accounted for. There are two main parts of droplet collision models for sprays - detecting colliding pairs of droplets and predicting the outcomes of these collisions. For the first part, we focus on the efficiency of the algorithm. We present an implementation of the arbitrary adaptive collision mesh model of Hou and Schmidt [1], and examine its efficiency in dealing with large simulations. Through theoretical analysis and numerical tests, we show that the computational cost of this model scales pseudo-linearly with respect to the number of parcels in the sprays. Regarding the second part, we examine the variations in existing phenomenological models used for predicting binary droplet collision outcomes. A quantitative accuracy metric is used to evaluate the models with respect to the experimental data set.

Advanced research in Spark-ignition (SI) engines has been focused on dilute-combustion concepts. For example, exhaust-gas recirculation is used to lower both fuel consumption and pollutant emissions while maintaining or enhancing engine performance, durability and reliability. These advancements achieve higher engine efficiency but may deteriorate combustion stability. One symptom of instability is a large cycle-to-cycle variation (CCV) in the in-cylinder flow and combustion metrics. Large-eddy simulation (LES) is a computational fluid dynamics (CFD) method that may be used to quantify CCV through numerical prediction of the turbulent flow and combustion processes in the engine over many engine cycles. In this study, we focus on evaluating the capability of LES to predict the in-cylinder flows and gas exchange processes in a motored SI engine installed with a transparent combustion chamber (TCC), comparing with recently published data.

Emissions from Diesel engines have been a major concern for many years, particularly with regards to the impact of NOx and particulate matter on human health. Exhaust gas re-circulation (EGR) is a widely used method in diesel engines for controlling NOx production. While EGR rates can be varied to ensure engine performance and reduce NOx emissions, EGR also influences the ignition delay, reduces the peak combustion temperature and increases particulate emissions. Moreover, the injection timing directly affects NOx and particulate emissions under the broad and highly variable operating conditions. An effective CFD-based design tool for diesel engines must therefore include robust and accurate predictive capabilities for combustion and pollutant formation, to address the complex design tradeoffs. The objective of the present study is to evaluate CFD modeling of diesel engine combustion and emissions for various combinations of EGR rates and injection timings.

Large-eddy simulation (LES) is a useful approach for the simulation of turbulent flow and combustion processes in internal combustion engines. This study employs the ANSYS Forte CFD package and explores several key and fundamental components of LES, namely, the subgrid-scale (SGS) turbulence models, the numerical schemes used to discretize the transport equations, and the computational mesh. The SGS turbulence models considered include the classic Smagorinsky model and a dynamic structure model. Two numerical schemes for momentum convection, quasi-second-order upwind (QSOU) and central difference (CD), were evaluated. The effects of different computational mesh sizes controlled by both fixed mesh refinement and a solution-adaptive mesh-refinement approach were studied and compared. The LES models are evaluated and validated against several flow configurations that are critical to engine flows, in particular, to fuel injection processes.

One of the best tools to explore complicated in-cylinder physics is computational fluid dynamics (CFD). In order to assess the accuracy and reliability of the CFD simulations, it is critical to perform validation studies over different engine operating conditions. Simulation-based design of SI engines requires predictive capabilities, where results do not need to be tuned for each operating condition. This requires the models adopted to simulate their respective engine physics to be reliable under a broad range of conditions. A detailed set of experimental data was obtained to validate the CFD predictions of SI engine combustion.

Gasoline Direct Injection (GDI) is a key technology in the automotive industry for improving fuel economy and performance of gasoline internal combustion engines. GDI engine performance and emission characteristics are mainly determined by the complex interaction of in-cylinder flow, mixture formation and subsequent combustion processes. In a GDI engine, mixture formation depends on spray characteristics. Spray evolution and mixture formation is critical to GDI engine operation. In this work, a multi-component surrogate fuel blend was used to represent the chemical and physical properties of the gasoline employed in the experimental engine tests. Multi-component spray models were also validated in this study against experimental spray injection measurements in a chamber. The spray-chamber data include spray-penetration lengths, transient spray velocities and droplet Sauter mean diameter (SMD) at different axial and radial distances from the spray tip, obtained using a PDPA system.

3-D Computational Fluid Dynamics (CFD) simulations have been performed using a detailed reaction mechanism to capture the combustion and emissions behavior of an IFP Energies nouvelles optical gasoline direct injection engine. Simulation results for in-cylinder soot volume fraction have been compared to experimental data provided by Pires da Cruz et al. [1] The engine was operated at low-load and tests were performed with parametric variations of the operating conditions including fuel injection timing, inlet temperature, and addition of fuel in the intake port. Full cycle simulations were performed including intake and exhaust ports, valve and piston motion. A Cartesian mesh was generated using automatic mesh generation in the FORTÉ CFD software. For the simulations, a 7-component surrogate blend was used to represent the chemical and physical properties of the European gasoline used in the engine tests.

While most published detailed reaction mechanisms for n-alkanes have been validated against shock-tube data that use pre-vaporized fuels, they have not been tested extensively using engine conditions. This is partly due to the complications of the effects of both spray and evaporation on ignition and on the gas-phase kinetics. In this study, CFD simulations of Ignition Quality Tests (IQT™) are used as a tool to validate the detailed reaction mechanisms, supplementing other validation tests that use more fundamental shock-tube data. The Ignition Quality Tester is a new ASTM standard for measuring the Cetane Number (CN) of fuels. Shock-tube data in the literature are limited for heavy n-alkanes of interest for engine fuels, which make CN data valuable for mechanism validation. The IQT employs a stationary combustion chamber that involves spray evaporation and mixing followed by combustion.

A multi-component fuel model is used to represent gasoline in computational fluid dynamics (CFD) simulations of a dual-fuel engine that combines premixed gasoline injection with diesel direct injection. The simulations employ detailed-kinetics mechanisms for both the gasoline and diesel surrogate fuels, through use of an advanced and efficient chemistry solver. The objective of this work is to elucidate kinetics effects of dual-fuel usage in Reactivity Controlled Compression Ignition (RCCI) combustion. The model is applied to simulate recent experiments on highly efficient RCCI engines. These engine experiments used a dual-fuel RCCI strategy with port-fuel-injection of gasoline and early-cycle, multiple injections of diesel fuel with a conventional diesel injector. The experiments showed that the US 2010 heavy-duty NO and soot emissions regulations were easily met without aftertreatment, while achieving greater than 50% net indicated thermal efficiency.

Designing advanced, clean and fuel-efficient engines requires detailed understanding of fuel chemistry. While knowledge of fuel combustion chemistry has grown rapidly in recent years, the representation of conventional fossil fuels in full detail is still intractable. A popular approach is to use a model-fuel or surrogate blend that can mimic various characteristics of a conventional fuel. Despite the use of surrogate blends, there remains a gap between detailed chemistry and its utilization in computational fluid dynamics (CFD), due to the prohibitive computational cost of using thousands of chemical species in large numbers of computational cells. This work presents a set of software tools that help to enable the use of detailed chemistry in representing conventional fuels in CFD simulation. The software tools include the Surrogate Blend Optimizer and a suite of automated mechanism reduction strategies.

To accurately predict emissions as well as combustion phasing in a homogeneous charge compression ignition (HCCI) engine, detailed chemistry needs to be used in Computational Fluid Dynamics (CFD) modeling. In this work, CFD simulations of an Oak Ridge National Laboratory (ORNL) gasoline HCCI engine have been performed with full coupling to detailed chemistry. Engine experiments using an E30 gasoline surrogate blend were performed at ORNL, which included measurements of several trace species in the exhaust gas. CFD modeling using a detailed mechanism for the same fuel composition used in the experiments was also performed. Comparisons between data and model are made over a range of intake temperatures. The (experiment & model) surrogate blend consists of 33 wt % ethanol, 8.7 % n-heptane and 58.3 % iso-octane. The data and simulations involve timing sweeps using intake temperature to control combustion phasing at a constant fuel rate.

Detailed knowledge of hydrocarbon fuel combustion chemistry has grown tremendously in recent years. However, the gap between detailed chemistry and computational fluid dynamics (CFD) remains, because of the high cost of solving detailed chemistry in a large number of computational cells. This paper presents the results of applying a suite of techniques aimed at closing this gap. The techniques include use of a surrogate blend optimizer and a guided mechanism reduction methodology, as well as advanced methods for efficiently and accurately coupling the pre-reduced kinetic models with the multidimensional transport equations. The advanced methods include dynamic adaptive chemistry (DAC) and dynamic cell clustering (DCC) algorithms.

To address the growing need for accurate predictions of combustion phasing and emissions for development of advanced engines, a more accurate definition of model fuels and their associated chemical-kinetics mechanisms are necessary. Wide variations in street fuels require a model-fuel blending methodology to allow simulation of fuel-specific characteristics, such as ignition timing, emissions, and fuel vaporization. We present a surrogate-blending technique that serves as a practical modeling tool for determination of surrogate blends specifically tailored to different real-fuel characteristics, with particular focus on model fuels for gasoline engine simulation. We start from a palette of potential model-fuel components that are based on the characteristic chemical classes present in real fuels. From this palette, components are combined into a surrogate-fuel blend to represent a real fuel with specific fuel properties.

In this paper, knock in a Ford single cylinder direct-injection spark-ignition (DISI) engine was modeled and investigated using the KIVA-3V code with a G-equation combustion model coupled with detailed chemical kinetics. The deflagrative turbulent flame propagation was described by the G-equation combustion model. A 22-species, 42-reaction iso-octane (iC8H18) mechanism was adopted to model the auto-ignition process of the gasoline/air/residual-gas mixture ahead of the flame front. The iso-octane mechanism was originally validated by ignition delay tests in a rapid compression machine. In this study, the mechanism was tested by comparing the simulated ignition delay time in a constant volume mesh with the values measured in a shock tube under different initial temperature, pressure and equivalence ratio conditions, and acceptable agreements were obtained.

A level set method (G-equation)-based combustion model incorporating detailed chemical kinetics has been developed and implemented in KIVA-3V for Spark-Ignition (SI) engine simulations for better predictions of fuel oxidation and pollutant formation. Detailed fuel oxidation mechanisms coupled with a reduced NOX mechanism are used to describe the chemical processes. The flame front in the spark kernel stage is tracked using the Discrete Particle Ignition Kernel (DPIK) model. In the G-equation model, it is assumed that after the flame front has passed, the mixture within the mean flame brush tends to local equilibrium. The subgrid-scale burnt/unburnt volumes of the flame containing cells are tracked for the primary heat release calculation. A progress variable concept is introduced into the turbulent flame speed correlation to account for the laminar to turbulent evolution of the spark kernel flame.