Modeling HPDI Natural Gas Heavy Duty Engine Combustion

Author(s):  
Guowei Li ◽  
Tim Lennox ◽  
Dale Goudie ◽  
Mark Dunn

CFD Modeling of the injection, the mixing, the combustion and the emission formation processes in a high pressure direct injection (HPDI) natural gas engine is presented in this paper. KIVA3V was used together with an injector model. Two sub-models had been developed that the concurrent injection, ignition and combustion of natural gas and diesel could be simulated. The gas injection was simulated with the injector model. In the injector model, the electromagnetism, the hydraulics and the mechanics were computed by solving a set of ordinary differential equations. Based on the engine experimental data, a combustion model was built in which premixed combustion of natural gas was excluded and the natural gas ignition was initiated by the pilot diesel combustion rather than a spontaneous process. The model calibration and validation are discussed. The model parameters were tuned against one set of engine test data. For the model validation, 30 engine test data were applied. The data were from HPDI engine tests at varied engine speeds, loads and injection timings with and without EGR. The model gave good agreement with the engine tests having no EGR. However, the model, in general, under-predicted the burning rate. With EGR, the model prediction errors were large and the NOx were under-predicted, though the trends were still captured.

Author(s):  
Stewart Xu Cheng ◽  
James S. Wallace

Direct injection natural gas (DING) engines offer the advantages of high thermal efficiency and high power output compared to spark ignition natural gas engines. Injected natural gas requires some form of ignition assist in order to ignite in the time available in a diesel engine combustion chamber. A glow plug — a heated surface — is one form of ignition assist. Simple experiments show that the thickness of the heat penetration layer of a glow plug is very small (≈10−5 m) within the time scale of the ignition preparation period (1–2 ms). Meanwhile, the theoretical analyses reveal that only a very thin layer of the surrounding gases (in micrometer scale) can be heated to high temperature to achieve spontaneous ignition. A discretized glow plug model and virtual gas sub-layer model have been developed for CFD modeling of glow plug ignition and combustion for DING diesel engines. In this paper, CFD modeling results are presented. The results were obtained using a KIVA3 code modified to include the above mentioned new developed models. Natural gas ignition over a bare glow plug was simulated. The results were validated against experiments. Simulation of natural gas ignition over a shielded glow plug was also carried out and the results illustrate the necessity of using a shield. This paper shows the success of the discretized glow plug model working together with the virtual gas sub-layer model for modeling glow plug assisted natural gas direct injection engines. The modeling can aid in the design of injection and ignition systems for glow plug assisted DING engines.


Author(s):  
Kang Pan ◽  
James S. Wallace

This paper presents a numerical study on fuel injection, ignition and combustion in a direct-injection natural gas (DING) engine with ignition assisted by a shielded glow plug (GP). The shield geometry is investigated by employing different sizes of elliptical shield opening and changing the position of the shield opening. The results simulated by KIVA-3V indicated that fuel ignition and combustion is very sensitive to the relative angle between the fuel injection and the shield opening, and the use of an elliptical opening for the glow plug shield can reduce ignition delay by 0.1∼0.2ms for several specific combinations of the injection angle and shield opening size, compared to a circular shield opening. In addition, the numerical results also revealed that the natural gas ignition and flame propagation will be delayed by lowering a circular shield opening from the fuel jet center plane, due to the blocking effect of the shield to the fuel mixture, and hence it will reduce the DING performance by causing a longer ignition delay.


Author(s):  
Kang Pan ◽  
James S. Wallace

A numerical study of ignition and combustion in a glow plug (GP) assisted direct-injection natural gas (DING) engine is presented in this paper. The glow plug is shielded and the shield design is an important part of the combustion system development. The results simulated by KIVA-3V indicated that the ignition delay (ID) predicted by an in-cylinder pressure rise was different from that based on a temperature rise, attributed to the additional time required to burn more fuel to obtain a detectable pressure rise in the combustion chamber. This time difference for the ignition delay estimation can be 0.5 ms, which is significant relative to an ignition delay value of less than 2 ms. To further evaluate the time difference between the two different methods of ignition delay determination, sensitivity studies were conducted by changing the glow plug temperature, and rotating the glow plug shield opening angle towards the fuel jets. The results indicated that the ID method time difference varied from 0.3 to 0.8 ms for different combustion chamber configurations. In addition, this study also investigated the influences of different glow plug shield parameters on the natural gas ignition and combustion characteristics, by modifying the air gap between the glow plug and its shield, and by changing the shield opening size. The computational results indicated that a bigger air gap inside the shield can delay gas ignition, and a smaller shield opening can block the flame propagation for some specific fuel injection angles.


Author(s):  
Luca Andreassi ◽  
Stefano Cordiner ◽  
Vincenzo Mulone ◽  
C. Reynolds ◽  
R. L. Evans

Compressed natural gas (CNG) has great potential as an alternative fuel for vehicle engines, and can reduce emissions and improve fuel economy. A single cylinder research engine has been modified to enable direct injection of a small quantity of fuel near the spark plug, independently of an overall lean homogeneous charge. Thus a partially stratified charge is formed within the chamber, which allows significant extension of the lean limit of combustion. This results in an improvement in specific fuel consumption. Numerical simulation also plays an important role in the development of such technological solutions. 3D simulations, in particular, are desirable to provide complete information about thermal and fluid dynamical fields within the chamber. In particular, among the developed numerical tools linked to the KIVA-3V code, special attention was dedicated to the formulation of the combustion model (CFM) turbulent combustion model based on the flamelet hypothesis), to adequately model non-homogeneities and lean mixture compositions. In this paper an optimization procedure is assessed, with the ultimate goal of designing combustion chambers properly devoted to be operated under lean (homogeneous and PSC) mixture conditions. The results related to the procedure definition and to its experimental validation are presented. Experimental and numerical data have been compared in terms of pressure cycles and heat release rate profiles. The overall results are encouraging, taking into special account the difficulty to reliably predict the key performance parameters without any “tuning interventions”, even when mixture richness and homogeneity were varied.


1993 ◽  
Vol 115 (4) ◽  
pp. 781-789 ◽  
Author(s):  
S.-C. Kong ◽  
R. D. Reitz

Ignition and combustion mechanisms in diesel engines were studied using the KIVA code, with modifications to the combustion, heat transfer, crevice flow, and spray models. A laminar-and-turbulent characteristic-time combustion model that has been used successfully for spark-ignited engine studies was extended to allow predictions of ignition and combustion in diesel engines. A more accurate prediction of ignition delay was achieved by using a multistep chemical kinetics model. The Shell knock model was implemented for this purpose and was found to be capable of predicting successfully the autoignition of homogeneous mixtures in a rapid compression machine and diesel spray ignition under engine conditions. The physical significance of the model parameters is discussed and the sensitivity of results to the model constants is assessed. The ignition kinetics model was also applied to simulate the ignition process in a Cummins diesel engine. The post-ignition combustion was simulated using both a single-step Arrhenius kinetics model and also the characteristic-time model to account for the energy release during the mixing-controlled combustion phase. The present model differs from that used in earlier multidimensional computations of diesel ignition in that it also includes state-of-the-art turbulence and spray atomization models. In addition, in this study the model predictions are compared to engine data. It is found that good levels of agreement with the experimental data are obtained using the multistep chemical kinetics model for diesel ignition modeling. However, further study is needed of the effects of turbulent mixing on post-ignition combustion.


Author(s):  
Ali Mohammadi ◽  
Masahiro Shioji ◽  
Yuki Matsui ◽  
Rintaro Kajiwara

Recently, an in-cylinder injection method has been considered for the improvement of thermal efficiency in natural-gas and hydrogen spark-ignition (SI) engines. However, the SI and combustion processes of gaseous jets are not well understood. The present study aims to provide fundamental data for the development of direct-injection SI gas engines. The ignition, combustion, and flame behavior of high-pressure and intermittent hydrogen and natural-gas jets in a constant volume combustion chamber were investigated. The effects of injection pressure, nozzle size, ambient pressure, and spark location were also investigated for various spark timings and equivalence ratios.


Author(s):  
Georg Fink ◽  
Michael Jud ◽  
Thomas Sattelmayer

Natural gas as an alternative fuel in engine applications substantially reduces both pollutant and greenhouse gas emissions. High pressure dual fuel (HPDF) direct injection of natural gas and diesel pilot has the potential to minimize methane slip from gas engines and increase the fuel flexibility, while retaining the high efficiency of a diesel engine. Speed and load variations as well as various strategies for emission reduction entail a wide range of different operating conditions. The influence of these operating conditions on the ignition and combustion process is investigated on a rapid compression expansion machine (RCEM). By combining simultaneous shadowgraphy (SG) and OH* imaging with heat release rate analysis, an improved understanding of the ignition and combustion process is established. At high temperatures and pressures, the reduced pilot ignition delay and lift-off length minimize the effect of natural gas jet entrainment on pilot mixture formation. A simple geometrical constraint was found to reflect the susceptibility for misfiring. At the same time, natural gas ignition is delayed by the early pilot ignition close to the injector tip. The shape of heat release is only marginally affected by the operating conditions and mainly determined by the degree of premixing at the time of gas jet ignition. Luminescence from the sooting natural gas flame is generally only detected after the flame extends across the whole gas jet at peak heat release rate. Termination of gas injection at this time was confirmed to effectively suppress soot formation, while a strongly sooting pilot seems to intensify soot formation within the natural gas jet.


Author(s):  
Michael Jud ◽  
Christoph Wieland ◽  
Georg Fink ◽  
Thomas Sattelmayer

An efficient computational fluid dynamics model for predicting high pressure dual-fuel combustion is one of the most essential steps in order to improve the concept, to reduce the number of experiments and to make the development process more coste-efficient. For Diesel and natural gas such a model developed by the authors is first used to analyze the combustion process with respect to turbulence chemistry interaction and to clarify the question whether the combustion process is limited by chemistry or the mixing process. On the basis of these findings a reduced reaction mechanism is developed in order to save up to 35% of computing time. The prediction capability of the modified combustion model is tested for different gas injection timings representing different degrees of premixing before ignition. Compared to experimental results from a rapid compression expansion machine, the shape of heat release rate, the ignition timing of the gas jet and the burnout are well predicted. Finally, misfiring observed at different geometric configurations in the experiment are analyzed with the model. It is identified that in these geometric configurations at low temperature levels the gas jet covers the preferred ignition region of the diesel jet. Since the model is based on the detailed chemistry approach, it can in future also be used for other fuel combinations or for predicting emissions.


Author(s):  
Kuo-Ting Hsia ◽  
Chad Garner ◽  
Gregory Vogel

The interface between rotating and stationary components in a gas turbine is commonly called “disk cavities”. In pursuit of a performance benefit from Secondary Air Flow System (SAF) flow reduction, rotating/stationary disk cavity flow ingestion management in the turbine section becomes an important subject. To completely avoid ingestion, large amount of purging flow is utilized which reduces engine performance. Too much ingestion and the risk of reduced engine durability becomes an issue. Therefore, a partial ingestion condition is desired for optimizing SAF for performance. Typical methodology would be analysing purging flow using 3-D computational fluid dynamics (CFD) modeling in conjunction with engine test data to validate the design and ensure optimal SAF partial ingestion. This approach puts a very heavy demand on computational resources even at steady state. To extend the practice into a transient and unsteady flow field to attempt full design optimization will be even more taxing on computational resources and analysis labor time. An quick alternative would be using a 1-D flow solver for the task in conjunction with baseload main gas path circumferential pressure variation either from main flow path 3D CFD or engine test data. Unfortunately, currently there are no 1-D flow solvers can resolve the poly-directional nature of partial flow ingestion. This paper presents a quick and reliable alternative using a numerical method for augmenting a 1-D flow solver to resolve the partial flow ingestion situation. This is used in conjunction with the 1-D flow solver to resolve the bulk cavity temperature and heat transfer in the disk cavity. The results of this augmented 1-D flow solver show excellent agreement to field measurement data and 3-D CFD solutions. The tool enables a very realistic transient thermal analysis with partial flow ingestion at a fraction of the cost and analysis time of a full 3-D CFD analysis. This enables a faster design optimization with multiple iteration of a turbomachinery disk cavity with partial flow ingestion.


Sign in / Sign up

Export Citation Format

Share Document