Search Results

In the present paper optimization tools are used to recommend low-emission engine combustion chamber designs, spray targeting and swirl ratio levels for a heavy-duty diesel engine operated at high-load. The study identifies aspects of the combustion and pollution formation that are affected by mixing processes, and offers guidance for better matching of the piston geometry with the spray plume geometry for enhanced mixing. By coupling a GA (genetic algorithm) with the KIVA-CFD code, and also by utilizing an automated grid generation technique, multi-objective optimizations with goals of low emissions and fuel economy were achieved. Three different multi-objective genetic algorithms including a Micro-Genetic Algorithm (μGA), a Nondominated Sorting Genetic Algorithm II (NSGA II) and an Adaptive Range Multi-Objective Genetic Algorithm (ARMOGA) were compared for conducting the optimization under the same conditions.

In-cylinder flow and combustion processes simulated with the standard k-ε turbulence model and with an alternative model-employing a non-linear, quadratic equation for the turbulent stresses-are contrasted for both motored and fired engine operation at two loads. For motored operation, the differences observed in the predictions of mean flow development are small and do not emerge until expansion. Larger differences are found in the spatial distribution and magnitude of turbulent kinetic energy. The non-linear model generally predicts lower energy levels and larger turbulent time scales. With fuel injection and combustion, significant differences in flow structure and in the spatial distribution of soot are predicted by the two models. The models also predict considerably different combustion efficiencies and NOx emissions.

A multi-objective genetic algorithm coupled with the KIVA3V release 2 code was used to optimize the piston bowl geometry, spray targeting, and swirl ratio levels of a high speed direct injected (HSDI) diesel engine for passenger cars. Three modes, which represent full-, mid-, and low-loads, were optimized separately. A non-dominated sorting genetic algorithm II (NSGA II) was used for the optimization. High throughput computing was conducted using the CONDOR software. An automated grid generator was used for efficient mesh generation with variable geometry parameters, including open and reentrant bowl designs. A series of new spray models featuring reduced mesh dependency were also integrated into the code. A characteristic-time combustion (CTC) model was used for the initial optimization for time savings. Model validation was performed by comparison with experiments for the baseline engine at full-, mid-, and low-load operating conditions.

An investigation of high speed direct injection (DI) compression ignition (CI) engine combustion fueled with gasoline (termed GDICI for Gasoline Direct-Injection Compression Ignition) in the low temperature combustion (LTC) regime is presented. As an aid to plan engine experiments at full load (16 bar IMEP, 2500 rev/min), exploration of operating conditions was first performed numerically employing a multi-dimensional CFD code, KIVA-ERC-Chemkin, that features improved sub-models and the Chemkin library. The oxidation chemistry of the fuel was calculated using a reduced mechanism for primary reference fuel combustion. Operation ranges of a light-duty diesel engine operating with GDICI combustion with constraints of combustion efficiency, noise level (pressure rise rate) and emissions were identified as functions of injection timings, exhaust gas recirculation rate and the fuel split ratio of double-pulse injections.

The influence of the constant C3, which multiplies the mean flow divergence term in the model equation for the turbulent kinetic energy dissipation, is examined in a motored diesel engine for three different swirl ratios and three different spatial locations. Predicted temporal histories of turbulence energy and its dissipation are compared with experimentally-derived estimates. A “best-fit” value of C3 = 1.75, with an approximate uncertainty of ±0.3 is found to minimize the error between the model predictions and the experiments. Using this best-fit value, model length scale behavior corresponds well with that of measured velocity-correlation integral scales during compression. During expansion, the model scale grows too rapidly. Restriction of the model assessment to the expansion stroke suggests that C3 = 0.9 is more appropriate during this period.

A computational study was performed to evaluate the effects of bowl geometry, fuel spray targeting and swirl ratio under highly diluted, low-temperature combustion conditions in a heavy-duty diesel engine. This study is used to examine aspects of low-temperature combustion that are affected by mixing processes and offers insight into the effect these processes have on emissions formation and oxidation. The foundation for this exploratory study stems from a large data set which was generated using a genetic algorithm optimization methodology. The main results suggest that an optimal combination of spray targeting, swirl ratio and bowl geometry exist to simultaneously minimize emissions formation and improve soot and CO oxidation rates. Spray targeting was found to have a significant impact on the emissions and fuel consumption performance, and was furthermore found to be the most influential design parameter explored in this study.

Engine design is a time consuming process in which many costly experimental tests are usually conducted. With increasing prediction ability of engine simulation tools, engine design aided by CFD software is being given more attention by both industry and academia. It is also of much interest to be able to use design information gained from an existing engine design of one size in the design of engines of other sizes to reduce design time and costs. Therefore it is important to study size-scaling relationships for engines over wide range of operating conditions. This paper presents CFD studies conducted for two production diesel engines - a light-duty GM-Fiat engine (0.5L displacement) and a heavy-duty Caterpillar engine (2.5L displacement). Previously developed scaling arguments, including an equal spray penetration scaling model and an extended, equal flame lift-off length scaling model were employed to explore the parametric scaling connections between the two engines.

A multi-objective genetic algorithm methodology was applied to a heavy-duty diesel engine at three different operating conditions of interest. Separate optimizations were performed over various fuel injection nozzle parameters, piston bowl geometries and swirl ratios (SR). Different beginning of injection (BOI) timings were considered in all optimizations. The objective of the optimizations was to find the best possible fuel economy, NOx, and soot emissions tradeoffs. The input parameter ranges were determined using design of experiment methodology. A non-dominated sorting genetic algorithm II (NSGA II) was used for the optimization. For the optimization of piston bowl geometry, an automated grid generator was used for efficient mesh generation with variable geometry parameters. The KIVA3V release 2 code with improved ERC sub-models was used. The characteristic time combustion (CTC) model was employed to improve computational efficiency.

The present work proposes a methodology for diesel engine development using multi-dimensional CFD and computer optimization. A multi-objective genetic algorithm coupled with the KIVA3V Release 2 code was used to optimize a high speed direct injection (HSDI) diesel engine for passenger car applications. The simulations were conducted using high-throughput computing with the CONDOR system. An automated grid generator was used for efficient mesh generation with 11 variable piston bowl geometry parameters. The first step in the procedure was to search for an optimal nozzle and piston bowl design. In this case, spray targeting, swirl ratio, and piston bowl shape were optimized separately for two full-load cases using simpler efficient combustion models (the characteristic time scale model and the shell ignition model). The optimal designs from the two optimizations were then validated using a combustion model with detailed chemistry (KIVA-CHEMKIN model and ERC n-heptane mechanism).

The recently developed KIVA-GA computer code was used in the current study to optimize the combustion chamber geometry of a heavy -duty diesel truck engine and a high-speed direct-injection (HSDI) small-bore diesel engine. KIVA-GA performs engine simulations within the framework of a genetic algorithm (GA) global optimization code. Design fitness was determined using a modified version of the KIVA-3V code, which calculates the spray, combustion, and emissions formation processes. The measure of design fitness includes NOx, unburned HC, and soot emissions, as well as fuel consumption. The simultaneous minimization of these factors was the ultimate goal. The KIVA-GA methodology was used to optimize the engine performance using nine input variables simultaneously. Three chamber geometry related variables were used along with six other variables, which were thought to have significant interaction with the chamber geometry.

A study of the combined use of split injections, EGR, and flexible boosting was conducted. Statistical optimization of the engine operating parameters was accomplished using a new response surface method. The objective of the study was to demonstrate the emissions and fuel consumption capabilities of a state-of-the-art heavy -duty diesel engine when using split injections, EGR, and flexible boosting over a wide range of engine operating conditions. Previous studies have indicated that multiple injections with EGR can provide substantial simultaneous reductions in emissions of particulate and NOx from heavy-duty diesel engines, but careful optimization of the operating parameters is necessary in order to receive the full benefit of these combustion control techniques. Similarly, boost has been shown to be an important parameter to optimize. During the experiments, an instrumented single-cylinder heavy -duty diesel engine was used.

Simultaneous two-component measurements of gas velocity and multi-dimensional numerical simulation are employed to characterize the evolution of the in-cylinder turbulent flow structure in a re-entrant bowl-in-piston engine under motored operation. The evolution of the mean flow field, turbulence energy, turbulent length scales, and the various terms contributing to the production of the turbulence energy are correlated and compared, with the objectives of clarifying the physical mechanisms and flow structures that dominate the turbulence production and of identifying the source of discrepancies between the measured and simulated turbulence fields. Additionally, the applicability of the linear turbulent stress modeling hypothesis employed in the k-ε model is assessed using the experimental mean flow gradients, turbulence energy, and length scales.

In-cylinder fluid velocity is measured in an optically accessible, fired HSDI engine at idle. The velocity field is also calculated, including the full induction stroke, using multi-dimensional fluid dynamics and combustion simulation models. A detailed comparison between the measured and calculated velocities is performed to validate the computed results and to gain a physical understanding of the flow evolution. Motored measurements are also presented, to clarify the effects of the fuel injection process and combustion on the velocity field evolution. The calculated mean in-cylinder angular momentum (swirl ratio) and mean flow structures prior to injection agree well with the measurements. Modification of the mean flow by fuel injection and combustion is also well captured.

Cycle-resolved analysis of velocity data obtained in the re-entrant bowl of a fired high-;speed, direct-injection diesel engine, demonstrates an unambiguous, approximately 100% increase in late-cycle turbulence levels over the levels measured during motored operation. Model predictions of the flow field, obtained employing RNG k-ε turbulence modeling in KIVA-3V, do not capture this increased turbulence. A combined experimental and computational approach is taken to identify the source of this turbulence. The results indicate that the dominant source of the increased turbulence is associated with the formation of an unstable distribution of mean angular momentum, characterized by a negative radial gradient. The importance of this source of flow turbulence has not previously been recognized for engine flows. The enhanced late-cycle turbulence is found to be very sensitive to the flow swirl level.

A micro-genetic algorithm (μGA) optimization method was applied to a heavy-duty, direct-injected diesel engine via an automated test bed system. The goal of this application was to demonstrate the feasibility and advantages of automated optimization experiments. With the genetic algorithm, no user input was required other than the factors of interest and their allowable ranges. This means that once the routine was initiated, it was essentially run undisturbed until a preset objective level was reached or a preset number of generations had been run. The automated μGA was successfully demonstrated at all points of the six-mode transient cycle simulation, excluding idle. To accomplish the automated experiments, an automated testing system was developed around a Caterpillar single-cylinder diesel engine.

The present study uses a numerical model to investigate the effects of flow turbulence on premixed iso-octane HCCI engine combustion. Different levels of in-cylinder turbulence are generated by using different piston geometries, namely a disc-shape versus a square-shape bowl. The numerical model is based on the KIVA code which is modified to use CHEMKIN as the chemistry solver. A detailed reaction mechanism is used to simulate the fuel chemistry. It is found that turbulence has significant effects on HCCI combustion. In the current engine setup, the main effect of turbulence is to affect the wall heat transfer, and hence to change the mixture temperature which, in turn, influences the ignition timing and combustion duration. The model also predicts that the combustion duration in the square bowl case is longer than that in the disc piston case which agrees with the measurements.

Simultaneous measurements of the radial and the tangential components of velocity are obtained in a high-speed, direct-injection diesel engine typical of automotive applications. Results are presented for engine operation with fuel injection, but without combustion, for three different swirl ratios and four injection pressures. With the mean and fluctuating velocities, the r-θ plane shear stress and the mean flow gradients are obtained. Longitudinal and transverse length scales are also estimated via Taylor's hypothesis. The flow is shown to be sufficiently homogeneous and stationary to obtain meaningful length scale estimates. Concurrently, the flow and injection processes are simulated with KIVA-3V employing a RNG k-ε turbulence model. The measured turbulent kinetic energy k, r-θ plane mean strain rates ( 〈Srθ〉, 〈Srr〉, and 〈Sθθ〉 ), deviatoric turbulent stresses , and the r-θ plane turbulence production terms are compared directly to the simulated results.

An optimization study combining multidimensional CFD modeling and a global, evolutionary search technique known as the Genetic Algorithm has been carried out. The subject of this study was a 2-stroke, spark-ignited, direct-injection, single-cylinder research engine (SCRE). The goal of the study was to optimize the part load operating parameters of the engine in order to achieve the lowest possible emissions, improved fuel economy, and reduced wall heat transfer. Parameters subject to permutation in this study were the start-of-injection (SOI) timing, injection duration, spark timing, fuel injection angle, dwell between injections, and the percentage of fuel mass in the first injection pulse. The study was comprised of three cases. All simulations were for a part load, intermediate-speed condition representing a transition operating regime between stratified charge and homogeneous charge operation.

The present work proposes a methodology for diesel engine development using scaling laws and computational optimization with multi-dimensional CFD tools. A previously optimized 450cc HSDI diesel engine was down-scaled to 400cc size using recently developed scaling laws. The scaling laws were validated by comparing the performance of these two engines, including pressure, HRR, peak and averaged temperature, and pollutant emissions. A novel optimization methodology, which is able to simultaneously optimize multiple operating conditions, was proposed. The method is based on multi-objective genetic algorithms, and was coupled with the KIVA3V Release 2 code to further optimize the down-scaled diesel engine. An adaptive multi-grid chemistry model was used in the KIVA3V code to reduce the computational cost of the optimization. The computations were conducted using high-throughput computing with the CONDOR system.

RNG k-ε closure turbulence dissipation equations are evaluated employing the CFD code KIVA-3V Release 2. The numerical evaluations start by considering simple jet flows, including incompressible air jets and compressible helium jets. The results show that the RNG closure turbulence model predicts lower jet tip penetration than the "standard" k-ε model, as well as being lower than experimental data. The reason is found to be that the turbulence kinetic energy is dissipated too slowly in the downstream region near the jet nozzle exit. In this case, the over-predicted R term in RNG model becomes a sink of dissipation in the ε-equation. As a second step, the RNG turbulence closure dissipation models are further tested in complex engine flows to compare against the measured evolution of turbulence kinetic energy, and an estimate of its dissipation rate, during both the compression and expansion processes.