Search Results

For various densities of gas jets including very light hydrogen and relatively heavy ones, the penetration length and diffusion process of a single high-speed gas fuel jet injected into air are computed by performing a large eddy simulation (LES) with fewer arbitrary constants applied for the unsteady three-dimensional compressible Navier-Stokes equation. In contrast, traditional ensemble models such as the Reynolds-averaged Navier-Stokes (RANS) equation have several arbitrary constants for fitting purposes. The cubic-interpolated pseudo-particle (CIP) method is employed for discretizing the nonlinear terms. Computations of single-component nitrogen and hydrogen jets were done under initial conditions of a fuel tank pressure of gas fuel = 10 MPa and back pressure of air = 3.5 MPa, i.e., the pressure level inside the combustion chamber after piston compression in the engine.

In our previous reports based on computations and fluid dynamic theory, we proposed a new compressive combustion principle for an inexpensive and relatively quiet engine reactor that has the potential to achieve thermal efficiency over 50% even for small combustion chambers having less than 100 cc. This can be achieved with colliding supermulti-jets that create complete air insulation to encase burned gas around the chamber center. We originally developed two small prototype engine systems for gasoline. First one with one rotary valve for pulsating intake flow and sixteen nozzles of jets colliding has no pistons. Next, we developed the second one having a strongly-asymmetric double piston system with the supermulti-jets colliding, although there are no poppet valves. The second prototype engine can vary point-compression strength due to the supermulti-jets and homogeneous compression level due to piston, by changing phase and size of two gears.

We have proposed the engine featuring a new compressive combustion principle based on pulsed supermulti-jets colliding through a focusing process in which the jets are injected from the chamber walls to the chamber center. This principle has the potential for achieving relatively silent high compression around the chamber center because autoignition occurs far from the chamber walls and also for stabilizing ignition due to this plug-less approach without heat loss on mechanical plugs including compulsory plasma ignition systems. Then, burned high temperature gas is encased by nearly complete air insulation, because the compressive flow shrinking in focusing process gets over expansion flow generated by combustion.

To improve urban air pollution, stringent emissions regulations for heavy-duty diesel engines have been proposed and will become effective in Japan, the EU, and the United States in a few years. To comply with such future regulations, it is critical to investigate the effects of intake and injection parameters and fuel properties on engine performance, efficiency and emissions characteristics, associated with the use of aftertreatment systems. An experimental study was carried out to identify such effects. In addition, the KIVA-3 code was used to gain insight into cylinder events. The results showed improvements in NOx-Smoke and BSFC trade-offs at high-pressure injection in conjunction with EGR and supercharging.

To suppress knock in small gasoline engines, the coolant flow of a single-cylinder engine was improved by using two methods: a multi-dimensional knock prediction method combining a Flamelet model with a simple chemical kinetics model, and a method for predicting combustion chamber wall temperature based on a thermal fluid calculation that coupled the engine coolant and the engine structure (engine head, cylinder block, and head gasket). Through these calculations as well as the measurement of wall temperatures and the analysis of combustion by experiments, the effects of wall temperature distribution and consequent unburnt gas temperature distribution on knock onset timing and location were examined. Furthermore, a study was made to develop a method for cooling the head side, which was more effective to suppress knock: the head gasket shape was modified to change the coolant flow and thereby improve the distribution of wall temperatures on the head side.

The objective of this article is to clarify the effect of thermal and equivalence ratio stratification on Homogeneous Charge Compression Ignition (HCCI) combustion under several conditions with three-dimensional computational fluid dynamics (3D CFD). Reynolds Averaged Navier-Stokes (RANS) simulation was used to calculate in-cylinder fluid dynamics. The 3D CFD simulation is also coupled with detailed chemical reaction to calculate HCCI combustion. First, the study with a simple engine model reveals that thermal stratification is more effective for prolonged combustion duration, which is a key factor for a high load limit of HCCI combustion, than equivalence ratio stratification. Thermal stratification has two-stage combustion: the combustion propagates from hot region slowly at first and then ignites in the entire in-cylinder region. Owing to this phenomenon, thermal stratification is more effective to mitigate HCCI combustion.

A high compression ratio is an effective means for improving the thermal efficiency of an internal combustion engines. However, a high compression ratio leads to a rapid rise in the combustion pressure, as it causes a high impulse force. The impulse force generates vibrations and noise by spreading in the engine. Therefore, reducing the vibration of the combustion (which increases as the compression ratio increases) can improve the thermal efficiency while using the same technology. We are conducting model-based research on technologies for reducing combustion vibration by applying a granular damper to a piston. To efficiently reduce the vibration, we suppress it directly with the piston, i.e., the source of the vibration. Thus, the damping effect is maximized within a minimized countermeasure range.

The purpose of this paper is to find a universal law to predict a turbulent burning velocity under various operating conditions and engine specifications. This paper presents the relationship between turbulent burning velocity and Karlovitz number. The turbulent burning velocity was measured using a single-cylinder gasoline engine, which has an external Exhaust Gas Recirculation (EGR) system. In the experiment, various engine operating parameters, e.g. engine speed and EGR rates, and various engine specifications, i.e. different types of intake ports were tested. Karlovitz number was calculated with Three Dimensional Computational Fluid Dynamics (3D-CFD) and detailed chemical reaction calculation, which condition was based on the experiment. The experimental and calculation results show that turbulent burning velocity is predicted by using Karlovitz number in the engine conditions, which varies depending on engine speed, EGR rates and the designs of intake ports.

In this study, reaction path analysis and modeling of NOx reduction phenomena by selective catalytic reduction (SCR) with NH3 over a Cu-chabazite catalyst were conducted considering changes in the valence state of Cu sites and local structure due to differences in ligands to the Cu sites. The analysis showed that in the Cu-chabazite catalyst, NOx was mainly reduced by adsorbed NH3 on divalent Cu sites accompanied by a change in valence state of Cu from divalent to monovalent. It is known that the activation energy of NOx reduction on a Cu-chabazite catalyst changes between low temperatures ≤ 200 °C and mid to high temperatures ≥ 300 °C. To express this phenomenon, a reversible hydrolysis reaction based on the difference in coordination state of hydroxyl groups (OH−) to Cu sites at low and high temperatures was introduced into the model.

In this study, reaction path analysis and modeling of NOx reduction phenomena by fast SCR reaction on a Cu-chabazite catalyst were conducted, considering the formation and decomposition of ammonium nitrate (NH4NO3). White crystals of NH4NO3 decompose at temperatures < 200 °C. Thus, the reaction behavior changes at 200 °C under fast SCR reaction conditions. NH4NO3 formation can occur on both Cu sites and Brønsted acid sites, which are active sites for NOx reduction in the Cu-chabazite catalyst, but it is unclear where NH4NO3 accumulates on the catalyst. Analyses using catalyst test pieces with different active sites were performed to estimate this accumulation. The results suggested that NH4NO3 accumulation does not depend on the presence of either Cu sites or Brønsted acid sites. Therefore, it is assumed that NH4NO3 can be accumulated everywhere on the catalyst, including on the zeolite framework. This phenomenon was included in the model as formation/accumulation sites S'.

In a Diesel Oxidation Catalyst (DOC) and Catalyzed Soot Filter (CSF) system, the DOC is used to oxidize additional fuel injected into the cylinder and/or exhaust pipe in order to increase the CSF's inlet temperature during soot regeneration. The catalyst's hydrocarbon (HC) oxidation performance is known to be strongly affected by the HC species present and the catalyst design. However, the engine operating conditions and additive fuel supply parameters also affect the oxidation performance of DOCs, but the effects of these variables have been insufficiently examined. Therefore, in this study, the oxidation performance of a DOC was examined in experiments in which both exhaust gas recirculation (EGR) levels and exhaust pipe injection parameters were varied. The results were then analyzed and optimal conditions were identified using modeFRONTIER.

A numerical study was carried out to investigate auto-ignition characteristics during HCCI predicted by using zero and multi-dimensional models combined with detailed kinetics including 116 chemical species and 689 elementary reactions involving iso-octane. In the simulation, homogeneous charge compression ignition of the fuel was analyzed under the same conditions as encountered in internal combustion engines. The results elucidated the combustible region and oxidation process of iso-octane with the formation and destruction of various chemical species in the cylinder.

To identify ways of achieving good mixture formation and heat release in diesel spray combustion, we have performed Large Eddy Simulation (LES) using a detailed chemical reaction mechanism to study the temporal and spatial distribution of the local equivalence ratios and heat release rate. Here we characterize the effect of the fuel injection rate profile on these processes in the combustion chamber of a diesel engine. Two injection rate profiles are considered: a standard (STD) profile, which is a typical modern common rail injection profile, and the inverse delta (IVD) profile, which has the potential to suppress rich mixture formation in the spray tip region. Experimental data indicate that the formation of such mixtures may extend the duration of the late combustion period and thus reduce thermal efficiency.

Natural gas is a promising alternative fuel for internal combustion engines because of its clean combustion characteristics and abundant reserves. However, it has several disadvantages due to its low energy density and low thermal efficiency at low loads. Thus, to assist efforts to improve the thermal efficiency of spark-ignited (SI) engines operating on natural gas and to minimize test procedures, a numerical simulation model was developed to predict and optimize the performance of a turbocharged test engine, considering flame propagation, occurrence of knock and ignition timing. The numerical results correlate well with empirical data, and show that increasing compression ratios and retarding the intake valve closing (IVC) timing relative to selected baseline conditions could effectively improve thermal efficiency. In addition, employing moderate EGR ratios is also effective for avoiding knock.

Natural gas has been used in spark-ignition (SI) engines of natural gas vehicles (NGVs) due to its resource availability and stable price compared to gasoline. It has the potential to reduce carbon monoxide emissions from the SI engines due to its high hydrogen-to-carbon ratio. However, short running distance is an issue of the NGVs. In this work, methodologies to improve the fuel economy of a heavy-duty commercial truck under the Japanese Heavy-Duty Driving Cycle (JE05) is proposed by numerical 1D-CFD modeling. The main objective is a comparative analysis to find an optimal fuel economy under three variable mechanisms, variable valve timing (VVT), variable valve actuation (VVA), and variable compression ratio (VCR). Experimental data are taken from a six-cylinder turbocharged SI engine fueled by city gas 13A. The 9.83 L production engine is a CR11 type with a multi-point injection system operated under a stoichiometric mixture.

In this study, fuel ignition timing parameters, in-cylinder pressure and heat release rates, and quantities of major exhaust gas emissions from a diesel engine were calculated using multi-dimensional CFD codes coupled with complex chemistry analysis. In addition, a sensitivity analysis of parameters was conducted to identify the major variables affecting these diesel combustion parameters. Firstly, diesel combustion analysis under typical operating conditions was carried out to validate the analytical methods used in the study, and then the effects of intake gas variables (e.g. temperature, and pressure) were investigated in detail in the sensitivity analysis. The results show that the main determinant of ignition timing in the engine is the spatial density of oxygen in the cylinder. This finding indicates that diesel combustion with high EGR and high boost pressure can provide both high thermal efficiency and low emissions.

Heavy soot deposition in wall-flow type diesel particulate filters reduces engine output and fuel efficiency. This necessitates forced regeneration to oxidize soot via exothermic reactions in a diesel oxidation catalyst upstream of the Diesel Particulate Filter (DPF). Soot loading in the wall of the DPF during forced regeneration causes much greater pressure drops than cake deposition, which is undesirable because high pressure drops reduce engine performance. We show that the description of soot deposition using a DPF model is improved by using a shrinking sphere soot oxidation sub-model. We then use this revised model to analyze cake deposition during forced regeneration, and to study the effects of varying the forced regeneration temperature and duration on the local soot reaction rate and soot mass distribution in the radial and longitudinal directions of the DPF channels during forced regeneration.

It has been recognized that alternative fuels such as liquid petroleum gas (LPG) has less polluting combustion characteristics than diesel fuel. Direct-injection stratified-charge combustion LPG engines with spark-ignition can potentially replace conventional diesel engines by achieving a more efficient combustion with less pollution. However, there are many unknowns regarding LPG spray mixture formation and combustion in the engine cylinder thus making the development of high-efficiency LPG engines difficult. In this study, LPG was injected into a high pressure and temperature atmosphere inside a constant volume chamber to reproduce the stratification processes in the engine cylinder. The spray was made to hit an impingement wall with a similar profile as a piston bowl. Spray images were taken using the Schlieren and laser induced fluorescence (LIF) method to analyze spray penetration and evaporation characteristics.

A variable valve timing (VVT) mechanism has been applied in a high-speed direct injection (HSDI) diesel engine. The effective compression ratio (εeff) was lowered by means of late intake valve closing (LIVC), while keeping the expansion ratio constant. Premixed charge compression ignition (PCCI) combustion, adopting the Miller-cycle, was experimentally realized and numerically analyzed. Significant improvements of NOx and soot emissions were achieved for a wide range of engine speeds and loads, frequently used in a transient mode test. The operating range of the Miller-PCCI combustion has been expanded up to an IMEP of 1.30 MPa.

Performance predictions of advanced turbocharged engines are becoming difficult because conventional engine models are built using performance map data of turbochargers with a proportional integral derivative (PID) controller. Improving prediction capabilities under transient test cycles or real driving conditions is a challenging task. This study applies a machine learning technique to predict turbocharger performances with high accuracy under steady-state and transient conditions. The manipulated signals of engine speed and torque created based on Compressed High-Intensity Radiated Pulse (Chirp signal) and Amplitude-modulated Pseudo-Random Binary Signal (APRBS) are used as inputs to the engine testbed. Data from the engine experiments are used as training data for the AI-based turbocharger model. High prediction accuracy of the AI turbocharger model is achieved with the co-efficient of determination in the model, and cross-validation results are higher than 0.8.