Valeri Golovitchev

Gasoline HCCI Modeling: An Engine Cycle Simulation Code with a Multi-Zone Combustion Model

For the application to Gasoline Homogenous Charge Compression Ignition (HCCI) modeling, a multi-zone model was developed. For this purpose, the detailed-chemistry code SENKIN from the CHEMKIN library was modified. In a previous paper, the authors explained how piston motion and a heat transfer model were implemented in the SENKIN code to make it applicable to engine modeling. The single-zone model developed was successfully implemented in the engine cycle simulation code AVL BOOSTTM. A multi-zone model, including a crevice volume, a quench layer and multiple core zones, is introduced here. A temperature distribution specified over these zones gives this model a wider range of application than the single-zone model, since fuel efficiency, emissions and heat release can now be predicted more accurately. The SENKIN-BOOST multi-zone model predictions are compared with experimental data. This demonstrates that the model can accurately predict fuel consumption, emissions and IMEP for a wide range of experimental operating conditions. An examination of the general trends predicted by the model (e.g. calculated fuel consumption, residual gas fraction and emissions as a function of IMEP) indicates that the proposed multi-zone model is a promising advance in Gasoline HCCI computer modeling.

Gasoline HCCI Modeling: Computer Program Combining Detailed Chemistry and Gas Exchange Processes

A skeletal reaction mechanism (101 species, 479 reactions) for a range of aliphatic hydrocarbons was constructed for application to computational fluid dynamics (CFD) Gasoline Homogeneous Charge Compression Ignition (HCCI) engine modeling. The mechanism is able to predict shock tube ignition delays and premixed flame propagation velocities for the following components: hydrogen (H2), methane (CH4), acetylene (C2H2), propane (C3H8), n-heptane (C7H16) and iso-octane (C8H18). The mechanism is integrated with a simulation code combining both modeling of detailed chemistry and gas exchange processes. This simulation tool was constructed by connecting the SENKIN code of the CHEMKIN library to the AVL BOOSTTM engine cycle simulation code. Using a complete engine cycle simulation code instead of a code that only considers the combustion process has a major advantage. The initial conditions at the intake valve closure (IVC) have no longer to be set. Typical initial conditions for a single-zone model are the average temperature of the mixture in the cylinder, the cylinder pressure and the species concentrations. When the engine cycle simulation code is used, the initial conditions consist of geometrical data for engine components and pressures and temperatures at different locations in the model scheme representing engine components connected by pipe elements. As gas exchange processes are included in the engine cycle simulation, the program will calculate the conditions at the IVC. The simulation program is used for parametric studies of the combustion process in the single-cylinder HCCI test engine at Chalmers University of Technology. Furthermore, provision for a link to future multidimensional CFD engine modeling is made. The model was matched to different test cases based on experimental data in order to obtain a tool giving accurate predictions for the different combinations of speed, load, excess air/fuel ratio and valve timings that are characteristic for the HCCI engine operation.

TOWARDS UNIVERSAL EDC-BASED COMBUSTION MODEL FOR COMPRESSION IGNITED ENGINE SIMULATIONS

The new EDC model formulation based on the operator-splitting procedure applied to the mass conservation equations for species participating in reversible chemical reactions which can be interpreted as representing combustion in a partially stirred reactor (PaSR) volume is presented. The model has been implemented in the KIVA-3V code, and examples of the model application to spray and gas combustion are illustrated, first, by the results of the 3-D modeling of the Diesel DI Volvo D12C engine. The combustion mechanism of diesel oil surrogate included of 68 species (including soot aromatic precursors) participating in 280 reversible reactions. Mean features of diesel spray engine combustion under conditions of delayed injection when auto-ignition has much in common with HCCI process are predicted in accordance with experimental data. As another application, the results of the 2-D simulations on a relatively coarse (∼1000 cells) grid for a combustion chamber of a natural gas fueled MHI HCCI model engine are presented and discussed. The reaction mechanism consisting of 45 species participating in 230 reactions has been used in the analysis. The calculated averaged pressure and temperature vs Ca histories were predicted in reasonable agreement with MHI experimental data.

Evaluation of Ignition Improvers for Methane Autoignition

The objective of this study was to estimate the efficiency of methane autoignition promotion by testing different ignition improvers including hydrogen, H2, hydrogen peroxide, H2O2, ozone,O3 and dimethyl ether, DME, (CH3)2O. This was accomplished by computing ignition delays for CH4/O2/Ar or N2 mixtures of various compositions, concentrations of the promoters, pressures and temperatures. Ignition delay times for additive-free mixtures were used for tuning a methane oxidation mechanism consisting of 185 reversible elementary reactions between 32 species. A selection of the reaction rate parameters available in the standard databases was made to optimize the agreement between simulation and experimental results for one particular set of test conditions (reference mixture) by refining the rate parameters of the most sensitive stages revealed by sensitivity analysis. The agreement achieved between model predictions and shock tube experimental data is very good. To investigate the effect of dimethyl ether on m...

CFD Analyses on 2-Stroke High Speed Diesel Engines

In recent years, interest has been growing in the 2-Stroke Diesel cycle, coupled to high speed engines. One of the most promising applications is on light aircraft piston engines, typically designed to provide a top brake power of 100-200 HP with a relatively low weight. The main advantage yielded by the 2-Stroke cycle is the possibility to achieve high power density at low crankshaft speed, allowing the propeller to be directly coupled to the engine, without a reduction drive. Furthermore, Diesel combustion is a good match for supercharging and it is expected to provide a superior fuel efficiency, in comparison to S.I. engines. However, the coupling of 2-Stroke cycle and Diesel combustion on small bore, high speed engines is quite complex, requiring a suitable support from CFD simulation. In this paper, a customized version of the KIVA-3v code (a CFD program for multidimensional analyses) has been used to address ports and combustion chamber design of a new project (a 3-cylinder 1.8L engine, with a power rating up to 150 HP). Multidimensional calculations have been supported by 1D engine cycle analyses, using GT-Power. Two types of combustion-scavenging system have been considered, both of them featuring direct injection: a configuration with exhaust poppet valves and another one with piston controlled ports. A development of both projects has been performed through a coupled 1d-3d computational approach. A first set of KIVA calculations has been performed, in order to characterize the scavenging and the port flow patterns of both configurations, considering three different operating conditions, representative an aircraft engine. Then, several combustion simulations have been run, for defining two chambers able to match the project goals (high fuel efficiency, limited in-cylinder peak-pressure). For the two best configurations, the most interesting calculation results are presented in the paper.

Chemical Model of Gasoline-Ethanol Blends for Internal Combustion Engine Applications

A semi-detailed chemical mechanism for combustion of gasoline-ethanol blends, which is based on sub-mechanisms of gasoline surrogate and for ethanol is developed and validated aiming at CFD engine modeling. The gasoline surrogate is composed of iso-octane, toluene, and n-heptane in volumetric proportions of 55%:35%:10%, respectively. In this way, the hydrogen-carbon atomic ratio H/C, which is around 1.87 for real gasoline, is accurately reproduced as well as a mixture equivalence ratio that is important for Gasoline Direct Injection engine applications. The integrated mechanism for gasoline-ethanol blends includes 120 species participating in 677 reactions. The mechanism is tested against experimental data on ignition delay times and laminar flame speeds, obtained for various n-heptane/iso-octane/toluene/ethanol-air mixtures under various equivalence ratios, initial temperatures, and pressures. Chemical, thermodynamic and transport properties used in the calculations are discussed.

Injection Strategy Optimization for a Light Duty DI Diesel Engine in Medium Load Conditions with High EGR rates

Further restrictions on NOx emissions and the expansion of current driving cycles for passenger car emission regulations to higher load operation in the near future (such as the US06 supplement to the FTP-75 driving cycle) requires attention to low emission combustion concepts in medium to high load regimes. One possibility to reduce NOx emissions is to increase the EGR rate. The combustion-temperature reducing effects of high EGR rates can significantly reduce NO formation, to the point where engine-out NOx emissions approach zero levels. However, engine-out soot emissions typically increase at high EGR levels, due to the reduced soot oxidation rates at reduced combustion temperatures and oxygen concentrations. The work presented in this paper focuses on the optimization of a triple injection strategy to study the effect of injection timing, fuel mass distribution over the different injections and fuel rail pressure on emissions, combustion noise and fuel consumption for operation at medium load (10 bar IMEP and upwards) and high EGR rates (41%). The results of some of the test cases are compared with those obtained from modelling in KIVA-3V. By using an optimized triple injection strategy, soot emission levels could be reduced to below 0.04 g/kWh and NOx emissions to below 0.4 g/kWh at a medium engine load of 10 bar IMEP in a single cylinder research engine.

CFD Modeling of Diesel Oil and DME Performance in a Two-Stroke Free Piston Engine

This paper presents the CFD KIVA3V, rel. 2 model and numerical results of combustion process simulations in a two-stroke, uniflow scavenging dual free piston engine, FPE, designed for electricity generation. Two fuels, diesel oil and dimethyl ether (DME), were studied in order to achieve HCCI-like combustion. Limited size reaction mechanisms were constructed and used in the modeling. The diesel oil surrogate mechanism consisting of a mixture of n-heptane and toluene includes 70 species participating in 305 reactions, whereas the mechanism for DME consists of 43 species participating in 211 reactions. The combustion development has been simulated in a FPE prototype geometry. It is illustrated that by varying the direct injection timings, a comparably efficient, low-emission operation has been achieved for both fuels. Specific difficulties when using slowly vaporizing diesel oil and rapidly vaporizing DME are outlined.

CFD Combustion and Emission Formation Modeling for a HSDI Diesel Engine Using Detailed Chemistry

In order to comply with current emissions regulations, a detailed analysis of the combustion and emission formation processes in the Diesel engines accounting for the effect of the main operating parameters is required. The present study is based both on 0D and 3D numerical simulations by compiling 0D chemical kinetics calculations for Diesel oil surrogate combustion and emission (soot, NOx) formation mechanisms to construct a φ-T (equivalence ratio - temperature) parametric map. In this map, the regions of emissions formation are depicted defining a possible optimal path between the regions by placing on the same map the engine operation conditions represented by the computational cells, whose parameters (equivalence ratio and temperature) are calculated by means of 3D engine modelling. Unlike previous approaches based on static parametric φ-T maps to analyze different combustion regimes and emission formations in Diesel engines, the present paper focuses on a construction of dynamic φ-T maps, in which the pressures and the elapsed times were taken in compliance with those calculated in the 3D engine simulations. The 0D chemical kinetics calculations have been performed by the SENKIN code of the Chemkin-2 library. In-cylinder conditions represented by computational cells with known φ and T are predicted using KIVA-3V code. When cells are plotted on the map, they identify the trajectories helping to navigate between the emissions regions by varying hardware and injection parameters. Sub-models of the KIVA-3V, rel. 2 code has been modified including spray atomization, droplet collision and evaporation, accounting for multi-component fuel vapor coupled with the improved versions of the chemistry/turbulence interaction model and new formulation of the combustion kinetics for the diesel oil surrogate (consisting in 70 species participating in 310 reactions). Simulations were performed for the HSDI 1.300 Fiat Diesel engine at optimized engine operating conditions and pilot injections. Finally, numerical results are compared with the experimental data on in-cylinder pressure, Rate of Heat Release, RoHR, and selected species distributions.Copyright

CFD-Based Optimization of a Diesel-fueled Free Piston Engine Prototype for Conventional and HCCI Combustion

This paper presents results of a parametric CFD modeling study of a prototype Free Piston Engine (FPE), designed for application in a series hybrid electric vehicle. Since the piston motion is governed by Newton’s second law, accounting for the forces acting on the piston/translator, i.e. friction forces, electrical forces, and in-cylinder gas forces, having a high-level control system is vital. The control system changes the electrical force applied during the stroke, thus obtaining the desired compression ratio. Identical control algorithms were implemented in a MATLAB/SIMULINK model to those applied in the prototype engine. The ignition delay and heat release data used in the MATLAB/SIMULINK model are predicted by the KIVA-3V CFD code which incorporates detailed chemical kinetics (305 reactions among 70 species). Since the piston motion and frequency, the rate of heat release and the initial in-cylinder conditions all affect each other, while predicted using different modelling tools with no direct coupling between them, an iterative procedure was used among models describing: 1. Piston dynamics governed by Newton’s second law including a high-level control system (using MATLAB/SIMULINK) 2. Combustion processes (using KIVA-3V) 3. Intake and exhaust system dynamics (using the GT-POWER module of the GT-SUITETM) Effects of varying parameters such as compression ratios, power supplied to the compressor, fuel injection timings and injection pressures have been studied in both conventional diesel and HCCI modes, the target being to identify optimal conditions for the combustion process in which the engine can be operated highly efficiently with very low-emissions.