CERC Annual Report 2011 - Chalmers · The three centers have a common board within SICEC, with a...

CERCCombustion Engine Research Center

Chalmers University of Technology

Annual Report 2011

CERC – Annual Report 20113CERC – Annual Report 2011 2

PrefaceThe year 2011 was a productive but mildly turbulent year for CERC. The three Swedish com-petence centers that perform research on combustion engines (CERC at Chalmers University of Technology, CCGEx at the Royal Institute of Technology in Stockholm, and KCFP at the University of Lund) continued to collaborate within a new structure called the Swedish Internal Combustion Engine Consortium (SICEC). SICEC meets three times per year in face-to-face meetings called the strategy and coordination group (SoS, based on the Swedish version of the name). SoS also meets over the web in between each face-to-face meeting. SoS’s func-tion is to coordinate research among the three centers, and to organize efforts to secure new funding and new industrial members.

The three centers have a common board within SICEC, with a common chair (Tommy Björkqvist). The 2011 CERC board has been made up of the following voting members:Tommy Björkqvist SICEC ChairAnna DuBois CERC /Chalmers Pär Gustafsson ABBBörje Grandin Volvo CarPer Lange Scania Lennart Skoogh SAAB Automobile Powertrain Sören Udd Volvo Powertrain

In addition, Bernt Gustafsson (from the Swedish Energy Agency), Ingemar Denbratt, Bo Egardt, Louise Olsson, and Monica Johansson (all from Chalmers) participate. The industrial members of the board represent companies that provide substantial support (e.g. at least 600 000 sek per year in cash and a similar amount in in-kind). Other signatories to the original CERC contract (some associated with just one project) include AB Volvo Penta, Hoerbiger Control Systems AB, Statoil A.S., and Lantmännen Aspen. SAAB Automobile has been through some upheavals and it appears they will be unable to provide financial support for 2011. Moreover, Honda R&D in Japan was hit hard by the earthquake and tsunami and they were unable to participate this year.

In 2011 eight PhD students and twelve senior researchers were involved in eleven projects. Stina Hemdal successfully defended her PhD thesis entitled “Optical Diagnostics Applied to Internal Combustion Engines: Catalysis, Sprays and Combustion”. Her opponent was Prof. Simone Hochgreb from Cambridge University in the UK. Chen Huang successfully defended her Licentiate Thesis as well, and Raúl Ochoterena took a job at SP in Borås. One project ended and a new project was started. As in previous years the CERC budget amounted to around 21 Msek, approximately half in cash and half in kind.

The Combustion division within the Institution for Applied Mechanics at Chalmers recently hired a new professor in computational fluid dynamics (CFD) into the Combustion Division. His name is Michael Oevermann and he is currently an Assistant Professor at the Freie Universität Berlin. His 1997 PhD is from the Technical University (RWTH) Aachen, Germany. Professor Oevermann will undoubtedly make significant contributions to CERC.

CERCs seminar day was held on Thursday, May 5. It was attended by the members of the international advisory board (IAB): Professor Simone Hochgreb, from Cambridge University in the UK; Dr. Mark Musculus from the Combustion Research Facility at Sandia National Laboratories in Livermore, California; and Professor Gianfranco Rizzo from the University of Salerno in Italy. The IAB also met with CERC researches and toured labs. In their reports they congratulated CERC for making significant improvements to the program, and gave some excellent advice for continued improvement.

Recently, Chalmers has committed to build a laboratory for research on hybrid electric vehicle powertrains (at a cost of around 15 MSEK), and the necessary modifications will begin soon. It will serve as a campus-wide resource for research on specialized combustion engines for hybrids and as range extenders for electric vehicles, and for research on hybrid systems and control. This will provide a clear link between CERC and the new national Swedish Hybrid Center (SHC), which is managed by Chalmers University. Moreover the Chalmers Area of Advance in Transportation has provided long term support for two post doctoral scholars who will make a strong connection between the Chalmers Competence Center for Catalysis (KCK) and CERC. They will focus on advanced catalytic exhaust gas cleanup technologies.

Combustion Engine Research Center (CERC)Table of Contents

2 General Background

3 Preface

4 Organization of CERC Research

7 Spray Guided Gasoline Direct-Injection

10 Light Duty Diesel Engine – Engine Control

14 Injection Strategies

18 Diesel Engine Optimisation

21 Alternative Fuels

24 Spray Turbulence Interaction

27 Combustion Models for Bio-Diesel Fuels

31 Modeling of Gasoline Direct-Injection Spark Ignition Engines

35 Optical Methods for Spray and Combustion Diagnostics

40 Human resources

42 Finances during the period 2010–2013

44 CERC Publications and Presentations during the period 2006–2011

General BackgroundChalmers’ Center of Excellence in internal combustion engines, called the Combustion Engine Research Center (CERC), was formally established on November 1, 1995 and inaugurated on March 26, 1996. At that time, the deci-sion to establish the Center was made by the Board of Chalmers University of Technology, based on an agreement between the Swedish Board for Technical and Industrial Development (NUTEK), Chalmers University of Technology, and a group of five Swedish companies. The agreement defined each party’s responsibilities with respect to financial commitments, scientific goals and use of research results. In 1997 the governmental coordination responsibilities were transferred to the Swedish National Energy Administration (STEM), which later changed its name to the Swedish Energy Agency (Energimyndigheten). The initial 10-year commitment ended at the close of 2005.

STEM and the industrial partners made early commitments to continue supporting CERC beyond the initial 10 year period. The formal application for continua-tion was thus approved for a new four year phase (2006–2009), which could be extended for another four years subject to satisfactory international evaluation in 2009. That evaluation by one panel of three scientists (who evaluated the research) and a second generalist panel (who evaluated the management and organization of CERC) was carried out in September of 2009 and CERC received a very posi-tive review. The decision to continue to support the Center has been made and we are now in the second four year phase which will last until the end of 2013, at which point another decision must be made.

During 2011, the following nine companies were full members of the Center:• ABB Automation Products AB• Saab Automobile Powertrain• Scania CV AB• Volvo Car Corporation AB• Volvo Powertrain AB• AB Volvo Penta• Hoerbiger Control System AB • Statoil AS

In addition, Aspen Petroleum AB and Reaction Design Inc. are members of CERC with single projects. Starting in 2010, the governing board consisted of one voting member from the academic community (with an additional four nonvoting members), five voting members from the participating companies, and one non-voting member from the Swedish Energy Agency (Energimyndigheten). The board chairmanship is now held by Tommy Björkqvist, who serves as a com-mon board chair for the three engine-related centers of competence in Sweden (CERC, CCGEx at the Royal Institute of Technology in Stockholm, and KCFP at the University of Lund). The three centers together are called the Swedish Internal Combustion Engine Consortium (SICEC), and Tommy is currently the director of SICEC.

CERC’s vision statement is: “Achieving sustainable powertrain technology through excellence in research and education”.

The mission statement provides some detail:• CERC will conduct groundbreaking engine and engine-related research with

a focus on turbulent combustion of transient sprays, engine efficiency, and emissions.

• Experiment and simulation will be strongly coupled; with the goal to develop successively more predictive engine models.

• CERC will educate top level engineers and scientists who will be able to secure rapid technological development for the Swedish motor industry.

• CERC will serve as a forum in which industrialists and academics can meet to exchange knowledge and information productively.

Mark Linne, Director,Combustion Engine Research Center

The coverPredicted in-cylinder temperature and droplet distribu-tion for Volvo D12C diesel engine fuelled by RME under 25% load and 25% EGR level.


CERC will certainly benefit immensely from both. Recent conversations with the Chalmers Area of Advance in Energy will probably lead to a similar arrangement related to hybrid vehi-cles. These initiatives are indicative of the exciting new developments underway at Chalmers and within CERC. 2012 is already shaping up to be an exciting year.

Our industrial partners and the Swedish Energy Agency have been highly supportive throughout 2011. They have been actively engaged in the Center and we thank them for that.

Mark LinneDirector, Combustion Engine Research [email protected]

Organization of CERC ResearchAs shown below, the ongoing research programs at CERC are supported by six scientific core competencies that can be found at Chalmers, based upon contributions from the departments of Applied Mechanics, Signals and Systems, and Chemical and Biological Engineering. Since the start of CERC the scientific foci have been on:• Spray modeling and measurements • Advanced studies of direct injection systems• Related combustion modes in engines• Engine control• Alternative fuels• Advanced measurement techniques

The 11 CERC research projects for 2011 fall within five main research areas, each led by one or two project leaders: 1. Petter Dahlander/Ingemar Denbratt for spark ignited and homogeneous charge compression

ignition engines (SI/HCCI – mostly using gasoline as fuel).2. Ingemar Denbratt/Tomas McKelvey for compression ignition (CI – mostly using diesel

fuel) engines and automatic control.3. Ingemar Denbratt for alternative engines and fuels.4. Valeri Golovitchev/Andrei Lipatnikov for engine modeling.5. Mats Andersson/Sven Andersson for optical diagnostics.

As mentioned, Raúl Ochoterena, who was leading the “Nanoparticles” project, took a job at SP in Borås. He had been mostly working on spray studies in Chalmers high pressure and temperature spray chamber to evaluate the same alternative fuels that Monika Johansson studied in a heavy duty Diesel engine for the Alternative Fuels project. That idea is certainly in keeping with the CERC strategy to couple spray vessel experiments to engine experiments and models as a way to learn more about the performance of a direct injected engine. For these reasons, the work Raúl did will be summarized in the Alternative Fuels report and there will be no Nanoparticles report this year. The project that was called Nanoparticles has now been

renamed “Spray fundamentals” and it will be led by Prof. Sven Andersson. We are currently searching for a new PhD student for that project.

The final project, in Waste Heat Recovery, is a common project funded by a consortium including several government agencies and industry. It is administered by Chalmers but shared with CCGEx at the Royal Institute and KCFP in Lund. It is not officially reported here in this document.

Each of the five research areas shown above has a reference group consisting of representatives from the interested CERC partners, and each reference group can encompass several projects. These reference groups meet roughly three times per year for in-depth discussions about research results and plans for the next phase. Note that reference groups 4 and 5 (Modeling and Diagnostics) cover enabling sciences, and so they are always blended with the other refer-ence groups in a way that maximizes their impact. The reference groups are the foundation for establishing project portfolios; projects are initiated, discussed and recommended to the SoS group via the reference groups, after which they are approved by the board. In some cases an area can include “associated” projects; those financed by sources outside of CERC (mainly the Swedish Energy Administration). Those are not shown in the diagram. The associated projects provide a real benefit to the center because they help CERC maintain critical mass in all five principal research areas.

Chalmers (via CERC) also hosts a national internal combustion engines research school, with participation by CCGEx at KTH and KCFP in Lund. The research school is headed by Sven Andersson and it is funded by VINNOVA (the Swedish funding agency for innovation and sustainable development).

The research strategy of CERC is discussed formally at each SoS meeting and more thor-oughly at CERC’s annual seminar. This document contains detailed descriptions of the CERC research projects carried out in 2011 and depicted in the figure above. Here we provide a short introduction to orient the reader:

1. SI/HCCI CombustionThe reference group for this research area included representatives from: SAAB Automobile Powertrain, Statoil, Volvo Car Corporation, and Chalmers University.Spray Guided Gasoline Direct-InjectionAim: Experimentally investigate the possibilities and limitations with a spray-guided gasoline direct injection combustion system using multi-hole and piezo actuated injectors. This project formally ended in 2010 but now has a continuation with a shifted focus for 4 more years. A new PhD student, Anders Johansson, has been hired.

2. Diesel Combustion and ControlThe reference group for this research area included representatives from: ABB, SAAB Automobile Powertrain, Hoerebiger, Scania, Volvo Car Corporation, Volvo Powertrain, and Chalmers University.Light Duty Diesel Engines for 2012 – CombustionAim: The primary goal of the project was to study effects of different injection, EGR and charging strategies using unconventional valve timings, compression ratios and injector configurations aiming at very low NOx and PM emissions at medium to high engine loads in order to understand how a “low emissions combustion concept” can be obtained at high engine load. This project ended near the beginning of 2011 and Arjan Helmantel has been on parental leave. Discussions about a continuation project are underway, but for now there is nothing new to report. Injection Strategies for Diesel EnginesAim: The main objective of this project is to minimize soot and NOx emissions without increases in BSFC. In order to achieve this, low temperature combustion (LTC) has been studied together with parameters concerning injection, fuel-air mixture and compression ratio, especially experiments related to the Miller cycle. Malin Ehleskog is an industrial PhD student from Volvo Powertrain who has been performing this work. She will defend her thesis in the winter of 2012 and the project will be adjusted to focus on new challenges.Light Duty Diesel for 2012 – Engine ControlAim: As an integrated part of the Light Duty Diesel Engine Project, to investigate the potential for closed-loop control by extracting and using information about the combustion in real-time

Figure 1. Organization of CERC research.


from crankshaft integrated torque sensor, ion current sensors and pressure sensors. The PhD student, Mikael Thor, will defend his thesis in the spring of 2012. Following that, ABB will leave the center. They are satisfied with the results of this project but do not see a market for their torque sensor in the motor industry. Diesel Engine OptimizationAim: The goal for this project is to develop a methodology that optimizes a multivariable problem with complicated response surface, and apply this method to the optimization of a diesel engine power train system with respect to emissions, fuel consumption, and human constraints such as noise, vibration and harshness (NVH). The accuracy of the optimization result will be assessed based on the accuracy of the underlying models. Markus Grahn, an industrial PhD student from Volvo Car, is working on this project.

3. Fischer-Tropsch fuels/Alternative fuelsThe reference group for this research area included representatives from: SAAB Automobile Powertrain, Scania, Statoil, Volvo Car Corporation, Volvo Technology, Volvo Powertrain, and Chalmers University.Fischer-Tropsch Fuels for Low Emissions in Diesel EnginesAim: Study Fischer-Tropsch fuels and compare their emission formations and combustion characteristics with conventional and other bio-fuels (e.g. RME). This project will end in 2012 when Monika Johansson defends her thesis.

4. Combustion modelingThe reference group for this research area included representatives from: SAAB Automobile Powertrain, Scania, Statoil, Volvo Car Corporation, Volvo Technology, Volvo Powertrain, Honda, and Chalmers University.Spray Turbulence InteractionAim: To improve the reliability of spray combustion modeling, focusing on spray/turbulence interaction and near-nozzle behavior (atomization). In practice this means investigating which turbulence properties are important for RANS modeling for capturing the transient injection. In the future, this work will also include LES modeling. Anne Kösters is the PhD student working on this project. She will defend her licentiate thesis in the winter of 2012. Combustion Models for Surrogate Bio-Diesel, RME, FuelAim: The object of the present project is to develop a full set of numerical spray combus-tion models for directed injection diesel engines fuelled by bio-fuels (RME) to evaluate the conditions in Volvo D12C and NED5 engines for effective low-emissions operation. Junfeng Yang, the PhD student on this project, will defend his thesis in the spring of 2012. This proj-ect will be re-defined with new leadership by Prof. Michael Oevermann because Prof. Valeri Golovichev will be retiring.Modeling of Gasoline Direct Injection Spark Ignition EnginesAim: The goal of this project is to develop models, methods, and a numerical platform for simulations of direct injection spark ignition engines that use different fuels (gasoline, ethanol, methanol and butanol, and mixtures of these fuels). The PhD student working on this project is Chen Huang.

5. DiagnosticsThe reference group for this research area included representatives from: SAAB Automobile Powertrain, Scania, Volvo Car Corporation, Volvo Technology, Volvo Powertrain, and Chalmers University.Advanced Laser-Based Methods for Spray ImagingAim: Develop and apply spray diagnostic methods for visualization of the distribution of fuel in vapor and liquid phase, quantification of fuel concentration and determination of temperature. This project interacts with other CERC-projects to: provide expertise and advice on spectros-copy and imaging and carry out collaborative studies for spray and combustion diagnostics.

In what follows we describe each project in much more detail, present the group of people who carry out this work, then present a budget for the year and end with a list of publications for the last year. Please note that only a selection of the results obtained in the various projects are presented here, with a focus on the most recent studies.

Much of what follows includes a number of specialized acronyms. For those who are not familiar, we define them here.

NomenclatureASOI after start of injectionATDC after top dead centerBSFC brake specific fuel consumptionBTDC before top dead centerCAD crank angle degreesCAP charge air pressureCFD computational fluid dynamicsDISC direct injection stratified chargeDISI direct injected spark ignitedEGR exhaust gas recirculationESC European stationary cycleFAME fatty acid methyl esterGDI gasoline direct injectionHC hydrocarbonsHCCI homogeneous charge compression ignitionIMEP indicated mean effective pressureLEV low emissions vehicle

LIF laser induced fluorescenceLIEF laser induced exciplex fluorescenceLTC low temperature combustionMEP mean effective pressureNEDC new European driving cycleNOx NO + NO2, not in any particular mixture ratioNOP needle opening pressurePDA phase Doppler anemometryPIV particle image velocimetryPLIF planar LIFPM particulate matterPPC partially premixed combustionRoHR rate of heat releaseSCR selective catalytic reduction (of NOx)SGDI spray guided direct injectionSULEV super ULEVULEV ultra low emissions vehicle

Research ProjectsSpray Guided Gasoline Direct-Injection

Objectives Lean burn combustion systems have the poten-tial to reduce CO2 emissions and for that reason this project seeks to develop them further. In the past, the SGDI project has concentrated on stratified combustion, but now the focus has been broadened to include homogenous lean combustion. The objective is to compare advan-tages and disadvantages between the two engine combustion modes with respect to fuel con-sumption, combustion stability, and emissions, (especially particulate numbers). Experiments investigate fuel spray characteristics using spray chambers and optically accessible, oper-ating engines. The aim has been to connect the spray shape to the resulting fuel-air mixture, combustion characteristics, and engine-out emissions.

The main project goals are:• To quantify particulate numbers under vari-

ous operating conditions and understand how they can be minimized.

• Quantify efficiencies of homogeneous and stratified lean burn operation, in comparison to homogeneous stoichiometric combustion, under realistic operating conditions.

BackgroundThe basic problems/challenges with a lean burn stratified system can be summarized as:1. Combustion robustness and stability2. Particulates number/mass3. NOx reduction at lean conditions 4. Combustion phasing that occurs too early

In a spray-guided combustion system the fuel injector and the spark plug are mounted close

Project leaderAssoc. Prof. Petter Dahlander

ResearchersAnders Johansson (PhD Student)Stina Hemdal (Post Doc)

to each other (in the so-called ‘close-spaced system’). The idea is that the spray itself should guide fuel/air mixture preparation, and the influence of gas and piston motion on the spray should be low. A precise spray with good atom-ization, low liquid fuel penetration, and low cycle-to-cycle variation is needed for reliable ignition. The two suitable types of fuel injectors are called outward opening piezo actuated and multihole solenoid actuated. Outward opening piezo injectors are used in the few spray-guided stratified combustion systems that are in pro-duction. Multihole injectors are less expensive and have improved substantially over the last few years, but there are currently no spray-guided lean burn GDI engines using multihole injectors in production.

New combustion systems require robust com-bustion and stable ignition. It is not clear how this can be achieved, especially with inex-pensive multi-hole injectors Ignition in spray-guided system depends strongly on the interac-tion between the spray event and the spark. For stratified combustion the local air-fuel ratio and flow velocity close to the spark at the time of ignition is important for stable combustion and misfire free operation [194,195].

Particulates are a problem especially during stratified operation since locally fuel rich areas promote particulate formation. This has led to new particulate legislation for SI engines; the first regulation was on particulate mass (2009) and in (2014) there will also be a regulation on the number of particulates. The legislation on particulate number is believed to be harder to meet than the legislation on mass [196].

Anders Johansson, Ph.D Student,Division of Combustion, Department of Applied Mechanics, Chalmers University of Technology.


Figure 2. The dummy piston mount-ed inside the spray vessel. The piston is tilted to achieve the same angle between the vertical axis of the spray and the piston as in the engine (since the injector is tilted when it is mounted in the engine).

Figure 3. Shadowgraph images of hollow cone sprays with 7.5 bar (left) and 20 bar (right) as back pressure.

Particulate filters may become necessary, but they are something industry wishes to avoid if possible. It is thought that particulate forma-tion is caused by fuel rich combustion owing to insufficient mixing time and long liquid fuel penetration. Higher fuel pressure increases air entrainment and mixing, which reduces fuel rich areas and particulate emissions [197]. On the other hand enhanced fuel penetration increases the risk of liquid fuel hitting the cyl-inder walls, the piston top or the spark plug which may lead to pool fires and increased soot emissions [198,199].

One of the drawbacks with lean burn is that NOx aftertreatment systems are complicated and expensive; hence the interest in keeping engine-out levels of NOx as low as possible. EGR, which is successfully used to reduce NOx emissions, could also reduce particle emissions from GDI engines [200,201].

Slow late burn in stratified mode requires early combustion phasing; earlier than optimum. One way to increase the slow late burn rate is to increase mixing, by increasing the fuel pressure [197] or by increasing tumble motion using a new cylinder head and or piston design.

In previous projects we have studied effects of parameters affecting the spray, for various types of injectors [51,64]. Parameters that were evaluated include: fuel pressure, type of fuel, fuel distribution (liquid/vapor), atomization, hole length/diameter, liquid fuel penetration, injector type, sprays under cold start conditions [110] etc. It has also been demonstrated in an optical engine that an outward opening piezo actuated injector can produce stable stratified combustion with very low fuel consumption [58]. However, this has only been demonstrated with fully open throttle and without EGR. Soot formation and oxidation during stratified combustion has also been studied in the opti-cal engine using an outward opening injector. [154].

This combustion system is complex and there are still many unanswered questions. It is believed that one can improve combustion (eg. combus-tion stability, EGR tolerance etc) by increasing the tumble motion since it increases mixing.

A study on the effect of a new cylinder head geometry on combustion would be beneficial. There are very few publications on particulate numbers describing what operating conditions can be problematic and whether particulate formation can be minimized. There is also no available study comparing particle numbers as a function of injector type. We realized that conclusions from spray chamber studies can be better linked to conclusions from engines studies if the boundary conditions are more engine-realistic. There are numerous spray chamber studies performed over a wide range of conditions described in the open literature, but for late injections only a limited number are carried out under realistic conditions (i.e. heated and at 15-20 bar back pressure).

Methods used for this yearThe experiments carried out this year include:1. Chamber studies of sprays with a dummy

piston (e.g. more engine-realistic boundary conditions) using high speed photography were conducted. The aim was to compare spray formation and liquid penetration from a multi-hole, solenoid-actuated injector and an outward-opening piezo-actuated injector at conditions equivalent to those under strati-fied operation in an engine. Here, engine realistic conditions also incorporate a dummy piston in the spray chamber with quartz glass plate to mimic the proximity of the piston crown to the spray.

2. Experiments in single cylinder engine using a new cylinder head with higher tumble motion are being set up. The aims are to mea-sure fuel consumption, emissions (including soot mass), particle diameters and particulate numbers at various air/fuel ratios, ie. homo-geneous lean, stoichiometric and stratified operation will be compared. Measurements should be performed at standard operating point (load/speed). Comparisons between different injectors with respect to particle numbers will be made. Sensors measuring the intake and exhaust pressures will be used estimates of internal EGR.

3. Experiments in optical engine under strati-fied combustion using a multi-hole injector are planned. In earlier experiments, the fuel distribution and combustion under stratified conditions were visualized inside the engine using an outward opening injector. The pur-pose is to compare the two types of injectors with respect to fuel distribution, soot forma-tion inside the cylinder, and engine out emis-sions including particulate numbers.

ResultsReferring to the numbers above, recent results include:1. Measurements in the spray chamber are fin-

ished and the results are under evaluation. A set of spray shadowgram images at selected time steps, obtained using an outward open-ing injector, are showed in Figure 3. The left hand column shows sprays obtained at 7.5 bar of backpressure and the right hand col-umn is for 20 bar (the two cases correspond to different start of injection times in an engine).

An adjustable dummy piston allows studies of the influence of the piston on spray forma-tion, and to study piston wetting by the spray. A set of measurements using the dummy piston at the correct distance from the injec-tor tip was performed.

The fuel energy content (heating value) injected was kept constant, necessitating dif-ferent injection durations. Individual spray events were captured using high speed video and the images were processed to obtain the liquid border and vertical penetration.

It is clear that the vortex formed by the hollow cone spray is affected by the increased back pressure. For the 7.5 bar case, the spray shape resembles results from earlier spray measure-ments performed at 6 bar [51], whereas the sprays obtained at 20 bar produce a much more compact fuel cloud. Around the end of injection (0.4 ms aSOI) one can see that the spray angle is roughly the same as in the 7.5 bar case; the vertical penetration is somewhat shorter, the vortices seem smaller and the spray is more narrow. It appears that the vor-tices are developed almost inside the bound-aries of the hollow cone instead of at the periphery. At 1 ms aSOI there are distinct differences between the sprays obtained at 7.5 and 20 bar. The fuel cloud at 20 bar is more compact with a larger amount of fuel in the center compared to the lower pressure case. Earlier LIF images of the fuel distribu-tion inside the single cylinder engine prior to ignition are shown in Figure 4. Here the fuel is visualized from below and a horizon-tal laser sheet has been used as illumination. At end of injection it can be seen that most of the fuel seems to be located in the center of the hollow cone and has not penetrated very far outside its borders, which is in agree-ment with the observations in the spray chamber for the higher pressure case. The results will be submitted to ICLASS, 2012.


2. The new cylinder head with increased tumble motion has been mounted together with mea-surement equipment. These experiments have not been finished and the results are expected early 2012. The experiments in the single cylinder engine will be finished during spring 2012 and the results will be submitted to SAE WC 2013.

3. Images acquired under stratified combustion using a multi-hole injector are currently under evaluation and the results will be pre-sented during 2012.

Conclusions• Fuel/air mixture preparation depends strongly

on back pressure. Under high back pressure liquid penetration is somewhat shorter and the recirculation zones are changed both in size and position.

• These spray experiments carried out at higher back pressures are generally in better agreement with engine experiments, com-pared to earlier spray studies performed at lower back pressures.

Other highlightsAnders Johansson has joined the project as a new PhD student during fall 2011.

Stina Hemdal defended her thesis March 11, 2011, and the faculty opponent was Simone Hochgreb.

Visiting student Pedro Marti from University of Valencia in Spain and Ph. D. student Thomas Rogers from the RMIT University in Australia have worked with the spray chamber experiments.

Figure 4. LIF images of the fuel distribution inside the single cylin-der research engine (SOI at 30 CAD bTDC) at the sparks electrodes (left) and 10 mm below the electrodes (right). 2 CAD corresponds to 0.164 ms., from [154].

Project leadersProf. Tomas McKelveyProf. Bo Egardt (Project co-leader)

ResearchersMikael Thor (PhD Student)Dr. Ingemar Andersson (Researcher)

Light Duty Diesel Engine – Engine Control

ObjectivesThis project investigates the use of a crank-shaft torque sensor for diesel engine control purposes. The objective is to determine the potential for using measurements from this sensor both for monitoring combustion and to control combustion in closed-loop. The work focuses on developing algorithms for combus-tion property estimation, e.g. estimation of start of combustion, combustion phasing and IMEP, and to demonstrate torque based closed-loop diesel engine combustion control online.

BackgroundA modern internal combustion engine is rapidly becoming a more and more complex system in response to increased demands on combus-tion efficiency and strict emission legislation.

Mikael Thor, PhD Student,Department of Signals and Systems,Chalmers University of Technology.

In order to meet these demands it is essential to optimize the performance of the engines by developing control systems that provide greater freedom for the engine to respond to chang-ing operating conditions. Traditionally, engine control systems consist of open-loop control strategies based on calibration maps for a fixed number of operating points. However, as the complexity of the engines increase so does the complexity of these maps and the resources needed to calibrate them are quickly becom-ing unacceptable [202]. As a result of this, the interest in closed-loop and adaptive combustion control systems has increased in recent years and these types of control systems appear to be essential for a further increase in internal combustion engine efficiency.

A closed-loop combustion control system relies on sensors that provide accurate and robust information about the combustion variables that are of interest to control. Such sensors may include in-cylinder pressure sensors, engine block accelerometers or crankshaft speed sen-sors. This project, however, focuses on informa-tion from a crankshaft torque sensor.

The torque sensorThe torque sensor used in this project, known as the Torductor-S, is developed by ABB and has been used in a number of different automo-tive applications in the past, see e.g. [203]. The sensor provides instantaneous high-resolution measurements of the crankshaft torque by detecting changes in the magnetic properties of the crankshaft caused by mechanical stress. In this project, the sensor is integrated as an extended part of the crankshaft between the last crank throw and the flywheel, see Figure5. As a result of this positioning, the measured torque signal contains torque contributions from all cylinders.

Previous workDuring the last decades, significant efforts have been made in the field of closed-loop combus-tion control. A majority of this work is based on measurements of in-cylinder pressure [204]. Except for the work that has been carried out at CERC, the available studies on the crankshaft torque sensor, e.g. for cylinder balancing [205] and engine monitoring in racing applications [206], have been performed by ABB in col-laboration with different partners. Recently, Lund University has also investigated the use of this type of sensor for HCCI combustion [207].

The work carried out in this project builds on the results from a previous CERC project, which ended in 2008, where torque based control of a spark-ignited engine was investigated [59,71]. While the overall control system structure from that project, shown in Figure 6, is left intact, the combustion model and crankshaft model have been modified in order to suit the diesel engine application in this project.

MethodThis project relies mainly on the use of experi-mental measurements to gather necessary data but simple zero-dimensional simulations are also used to a small extent. The experimen-tal engine is based on a standard production 5-cylinder 2.4-liter diesel engine from Volvo Car Corporation. However, modifications have been made to the engine’s crankshaft in order to fit the torque sensor and all cylinders have been instrumented with pressure sensors that are used for reference measurements. The engine can be controlled using the standard production control system but this system does

not offer enough freedom to be able to dem-onstrate torque based closed-loop combustion control online. An open-source control system is therefore being developed throughout the course of this project. This control system is based on hardware and software from National Instruments and Drivven and offers complete freedom for all engine actuators as well as the possibility to handle experimental sensors.

The experiments are designed to collect in-cyl-inder pressure reference data along with data from the investigated torque sensor and other standard sensors in different engine operating points and using different injection strategies. After this, the collected data is analyzed offline using system identification methods in order to create the models and algorithms needed for the combustion property estimation.

ResultsPreviously, this project has reported results on combustion and crankshaft modeling. Using the crankshaft model alone delivers cylinder indi-vidual IMEP estimates with a bias of less than 3% for all cylinders [93]. When using the crank-shaft model together with a combustion model based on Vibe functions, the cylinder individual combustion phasing is estimated with a bias of less than 0.6 CAD for all cylinders [104].

Figure 5. An illustration of a crankshaft with a torque sensor installed between the final crank throw and the flywheel. The illustration is a courtesy of Stefan Larsson.

Figure 6. The structure of a closed-loop combustion control system based on measurements of the crankshaft torque.


During the past year, the work in the project has focused on the recently developed combus-tion net torque concept, online implementation for demonstration of torque based combustion phasing control, and studies of ideal torque sig-nal properties.

Combustion phasing estimation based on combustion net torquePreliminary results regarding the concept of combustion net torque were reported already last year and the study has continued during 2011. Combustion net torque, illustrated in Figure 7, is defined as the difference between the torque produced by a fired engine cycle and the torque produced by a corresponding motored engine cycle. Hence, combustion net torque describes the torque contribution added by the combustion event. The work during the past year has led to two different combustion property estimation methods based on combus-tion net torque, one that estimates the entire burned mass fraction trace by solving a convex optimization problem and another that directly estimates the location of 50% burned mass frac-tion through a nonlinear black-box mapping. Compared with each other, the method for estimation of the entire burned mass fraction trace suffers more from errors in the cylinder separation. Therefore, the direct combustion phasing estimation method has been chosen for use in the online evaluations of torque based combustion phasing control. The combustion phasing estimation error distributions for these two methods, based on data from the crankshaft torque sensor, are depicted in Figure 8.

Torque based combustion phasing controlPreliminary experiments studying closed-loop combustion phasing control based on the previ-ously mentioned combustion net torque have been conducted during the latter part of 2011. The results from these preliminary measure-ments, where disturbances in the fuel injection timing and EGR flow are detected and counter-acted, are illustrated in Figure 9 and Figure 10. Even though some algorithm refinements are needed, the results look promising. One large performance limitation is the hardware process-ing capabilities that only allow the estimates to be updated every 4-5 cycles at 1500 RPM.

Figure 8. Combustion phasing estimation errors using the method for entire burned mass fraction estimation (left) and the method for direct combustion phasing estimation (right). The results are based on data from the crankshaft torque sensor.

Figure 11. Examples of true (solid) and recreated (dashed) cylinder pressure traces. In some cases, combustion information close to TDC can be difficult to pick up based on torque data.

Figure 9. Preliminary results of torque based combustion phasing control un-der the influence of injection timing disturbances. The upper plot shows the lo-cation of 50% burned mass fraction without closed-loop control while the lower plot illustrates the counteracting effect achieved with the controller active.

Figure 7. An illustration of combustion net torque, defined as the difference between fired and motored torque.

The combustion information content of a torque signal The work in the project also revisited the parameterized combustion models that have been investigated previously in connection with the torque ratio concept. This time these mod-els, based on Vibe functions [208], were used together with ideal torque data, i.e. torque calcu-lated from cylinder pressure, in order to assess how much combustion information it is possible to find in a torque signal in the ideal case. Hence, this study provides an indication of the upper performance bound for torque based combus-tion property estimation. The parameterized models are fitted to torque domain data and later used to recreate the corresponding cylinder pressure trace. The recreated cylinder pressure trace is then used in order to estimate a vari-ety of combustion properties. The results show that ideal estimation of combustion phasing and the maximum cylinder pressure is robust for most cases while the ideal estimation of start of combustion and the maximum cylinder pres-sure gradient is more sensitive to disturbances. Generally, the estimation uncertainties tend to increase close to TDC where the signal-to-noise ratio of a torque signal is low. This is also illus-trated in Figure 11, where examples of recreated cylinder pressure traces are shown.

Conclusions and ongoing workThe work during the past year has focused on three areas. First, a greater understanding of the properties of combustion net torque has been obtained. This resulted in two new methods for combustion property estimation, one that estimates the entire burned mass fraction trace and another that estimates combustion phasing directly. Secondly, preliminary experiments for the demonstration of torque based combustion phasing control have been performed. The gath-ered data will be analyzed and used for improv-ing the current performance of the estimation algorithms. Finally, conducted studies of ideal torque data have led to in depth knowledge of the properties of a torque signal and the con-nection between different torque measures and the burned mass fraction.

The project will, during early 2012, focus on preparing and conducting the final experiments needed in order fully demonstrate torque based combustion phasing control online. The plan is then to conclude the project by writing the final publications and for Mikael Thor to defend his Ph.D. thesis before the summer of 2012.

Figure 10. Preliminary results of torque based combustion phasing control under the influence of EGR flow disturbances. The upper plot shows the loca-tion of 50% burned mass fraction without closed-loop control while the lower plot illustrates the counteracting effect achieved with the controller active.


Project leaderProf. Ingemar Denbratt

ResearcherMalin Ehleskog (PhD Student)

Injection Strategies

Objectives Increasing public awareness of our effect on the environment together with rising fuel costs and finite oil reserves has prompted the introduction of stringent legislation on engine emissions. Furthermore, legislation on the fuel efficiency of Heavy Duty trucks is expected to be introduced in the near future. It is therefore increasingly important to reduce engine-out emissions while maintaining low fuel consump-tion. Previous research has shown that HCCI or low temperature combustion with high levels of EGR can reduce both soot and NOx emis-sions. However, these combustion concepts also generate high CO and HC emissions and have high fuel consumption. Moreover, they are only viable for low loads.

The objective of the project, therefore, is to investigate the scope for reducing HC and CO emissions during low temperature combustion in order to improve fuel economy. The poten-tial for using this combustion mode under higher loads was also investigated. A final line of investigation focused on using partially premixed combustion as a means of reducing soot and NOx emissions without increasing fuel consumption.

A variety of injection strategies has been inves-tigated with the aim of improving the mixing of fuel and air in the cylinder and thereby reducing CO and HC emissions while promoting soot oxidation. Other important studies have focused on the effect of the charge air pressure and EGR on the formation of emissions and the scope for using a variable valve actuation (VVA) sys-tem to vary the effective compression ratio and thereby adjust the temperature and pressure of the charge air.

Background The first years of this project were dedicated to Low Temperature Combustion (LTC) using very high levels of EGR. Low temperature combustion, with very low soot and NOx emis-sions, was achieved using a standard compres-sion ratio of 18.5 at 25% load. However, it was found that a lower compression ratio was nec-essary to achieve higher loads. The HC and CO emissions were extremely high at the EGR levels where LTC took place. These unburned emissions could be reduced by optimized SOI and increased injection pressure but the fuel consumption remained very high.

The results, however, revealed an area of inter-mediate EGR levels where the NOx emissions had decreased substantially but the soot emis-sions had not yet increased to high levels, and where CO and HC emissions still were very

Malin Ehleskog, Ph.D StudentDivision of Combustion, Department of Applied Mechanics, Chalmers University of Technology.

low. This partially premixed form of combus-tion gave acceptable levels of soot and NOx emissions together with low fuel consumption. To further improve the engine out emissions, it would be necessary to investigate the effect that different engine parameters have on combus-tion, soot formation and soot oxidation. The potential for using various valve timing strate-gies (with or without EGR) to further reduce emissions, without impairing fuel consump-tion, has been explored. By closing the inlet valve early, during the intake stroke (EIVC), or late during the compression stroke (LIVC) the effective compression ratio can be reduced, thus changing the pressure and temperature before combustion.

De Ojeda [209] used EIVC to lower the pressure and temperature during the compression stroke which produced longer ignition delay caused by more homogenous mixing. He investigated the effects of varying IVC from -230 to -130 atdc (baseline), with NOx emissions and CA50 kept constant by varying the SOI and EGR. Charge air pressure was also kept relatively constant. EIVC reduced the charge mass, the effective compression stroke and compression pressure. The maximum combustion pressure was not reduced as much owing to more rapid combustion with EIVC. EIVC was also found to have a cooling effect on the charge mass, reducing the temperature at TDC by 100 K. Exhaust temperatures were raised by over 100 K with early IVC [209].

De Ojeda found that EIVC dramatically reduced soot emissions at constant NOx while also reduc-ing the fuel consumption. The reduced mass air flow did not have a detrimental effect on soot emissions. The amount of EGR needed to maintain low NOx emissions decreased, but the O2 concentration in the intake remained almost the same. CO and HC emissions increased as IVC was advanced. The global equivalence ratio increased, but owing to the longer igni-tion delay, soot emissions remained low owing to lower local equivalence ratio. In addition, the premixed burn did not tend to soot heav-ily, thus the soot emissions decreased with the amount of fuel that burned in premixed mode. Soot emissions and ignition delay were found to be dependent on IVC, but bsfc appeared to be more dependent on combustion phasing [209].

Retarded IVC has been more intensively inves-tigated than early IVC strategies. Murata et al achieved [211] simultaneous reductions in NOx and soot emissions by using LIVC together with 40% EGR. When early injection and LIVC were used in combination, soot reductions were

achieved despite a low excess air ratio, indicating that prolonging the premixing time has a stronger effect on soot than does lowering the excess air ratio. Unfortunately, HC and CO emissions were increased, resulting in a fuel consumption pen-alty with early injection and LIVC. The exhaust temperature was also increased [211].

Computational analysis reported in the same publication showed that early injection and LIVC produced the longest ignition delay and the most homogenous lean mixtures at equiva-lence ratios below 2.5. This combination also yielded spatially wider heat generation regions and lower combustion temperatures (below 2100 K), despite the leanness of the fuel-air mixtures, thus avoiding both local over-mixing and high temperature regions [211].

Partially premixed combustion offers the poten-tial to reduce both soot and NOx emissions with-out large CO and HC penalties (and hence fuel consumption penalties). With the help of a VVA system, the effective compression ratio can be reduced, potentially enabling even lower soot emissions and fuel consumption.

MeasurementsA measurement campaign was performed on a single cylinder heavy duty diesel engine based on Volvo D12C, see Figure 12. The en-gine was equipped with a unit injector and the

compression ratio was lowered to 17. The inlet valves were controlled using a pneumatic vari-able valve system manufactured by Cargine Engineering AB, Sweden. The system enables the valve timing, opening duration, and valve lift to be controlled. The valves are operated using pressurized air (6 bar) controlled by two solenoids. A signal to the first solenoid fills the actuator with pressurized air, determining the opening timing. The length of this signal also determines the opening duration of the valve. A signal to the second solenoid is used to stop the filling of the actuator with compressed air, thus controlling the valve lift. During the open-ing duration, the valve is held at the desired lift height by hydraulic pressure in the actuator.

Because the valves open and close very rapidly, the IVO time must be retarded in comparison to standard opening timings, to avoid contact between the valve and the piston. IVO was set at 30 CAD aTDC compared to 10 CAD bTDC which is standard for this engine. The standard IVC is 30 aBDC for this engine. The open-ing and closing of the valves was monitored by an optical sensor and the valve lift profile was logged by an inductive sensor. No stable intermediate valve lift values were possible to achieve in this investigation, so the valve lift was always held at the maximum.

The effect of IVC–timing was investigated for three early and three late IVC timings at 50 and 25% load, see Figure 13. The baseline (IVC 0) was set at 540 CAD (BDC). The IVC was then advanced by 30, 50 and 70 CAD with the

Figure 12. The AVL single-cylinder engine equipped with the Cargine variable valve system.

Figure 13. Normalised valve lift cur-ves for the investigated IVC timings.

Figure 14. The effective compres-sion ratio for the various IVC timings investigated.


IVO timing kept constant, thus shortening the duration. The IVC was also retarded to 60, 80 and 100 CAD after baseline which results in a longer opening duration. The resulting effec-tive compression ratio can be seen in Figure 14. The charge air pressure was adjusted to keep the airflow through the engine constant for a given EGR level when the IVC was varied. This enabled constant equivalence ratio and constant oxygen concentration for a given EGR level. Furthermore, at 25% load, the effect of increased swirl was investigated. By deactivat-ing one of the valves, the swirl number was increased from 0.2 to 2.5 for the baseline IVC of the engine. In this case the charge air pressure was also adjusted to maintain constant airflow. The CA50 of the apparent heat release was kept constant with varied SOI.

Speed [rpm] Load Torque

[Nm]EGR [%]

Airflow [kg/h]

CA50[CAD aTDC]

1200 50% 146 0% 200 9.5 ± 0.5

30% 94 10.6 ± 0.3

25% 74 0% 148 8.6 ± 0.7

50% 64 9.1 ± 0.7

w. swirl 0% 148 9.2 ± 0.5

Results The effect of IVC on pressure and temperature at TDC was small for the cases without EGR, see Figure 15. This is believed to be a result of the high airflow and high charge air pressure needed to maintain it. When the airflow was reduced with EGR, the effect of the reduced compression ratio on pressure and temperature increased. Both advanced and retarded IVC means that the charge air pressure has to be increased in order to maintain a constant airflow

Early IVCFor the cases without EGR at 50% load, there were small differences in pressure and tem-perature at TDC for the different inlet valve closings. The ignition delay was prolonged, enabling more premixed combustion and the NOx emissions were reduced, see Figure 16. Since the oxygen concentration and charge mass were constant, the NOx emission reduction is believed to be a result of the small reduction in temperature and pressure at TDC, which increases the ignition delay.

For early IVC with EGR both the tempera-ture and the pressure were reduced at TDC (Figure 15) as a result of the lower effective compression ratio. This is in agreement with the results previously reported by De Ojeda [209]. As expected, this change will prolong the ignition delay enabling more of the combus-tion to occur in premixed mode (Figure 17). When 30% EGR was used, the soot emissions increased with advanced IVC, see Figure 16. However, there was a slight increase in equiva-lence ratio for these IVC timings. Furthermore, the in-cylinder temperature is lower (Figure 18). This is believed to worsen oxidation and it leads to increased CO (Figure 19) and HC emissions. The increased pumping losses due to the higher charge air pressure and accompany-ing back pressure and the increase in unburned emissions also increased the fuel consumption for the very early IVC cases (Figure 19).

The results obtained with EGR and a constant airflow matches the results obtained by Benajes et al [211] for constant oxygen concentration with varied EGR. However, the study presented in this paper shows larger decrease in in-cyl-inder temperature than Benajes found [212].

The effect of early IVC was also investigated at 25% load. The results followed the same trend as for 50% load for the low EGR cases but when 50% EGR was used, advanced IVC resulted in reduction of soot emissions for very low NOx emissions, see Figure 20. Since the equivalence ratio was kept constant, this appears to be due to the lower effective compression ratio resulting in a longer ignition delay. There was a slight increase of CO emissions but the fuel consump-tion remained low.

Thus, the prolonged ignition delay and increased premixing that the reduced effective compres-sion ratio gives because of early IVC can give both low soot and NOx emissions without increased CO emissions or fuel consumption if both air flow and CA50 are kept constant. However, the expected extra cooling of the charge air during the expansion in the intake stroke appears to be counteracted by increased heat transfer from the walls. Thus, the tempera-ture reduction at TDC is similar for early and late IVC.

Late IVCSimilar to the results obtained for early IVC, there are small differences in pressure and temperature at TDC for the cases without EGR when the IVC was delayed. The ignition delay was prolonged, enabling more premixed com-bustion. Together with similar or reduced in-cylinder temperatures for late IVC, this reduces the NOx emissions, see Figure 16.

When EGR was added, the advanced IVC and increased amount of premixing produced

reduced soot emissions (Figure 16). The in-creased premixing and more rapid combustion with the later IVC (Figure 17) improved oxida-tion, resulting in low CO emissions (Figure 19) and increased exhaust temperatures. However, the increased pumping losses together with less combustion at constant volume despite con-stant CA50 deteriorated the fuel consumption (Figure 19). The fuel consumption for late IVC was lower than for the early IVC at comparable compression ratios. This is believed to be due to the fact that equivalence ratio was maintained together with the low CO emissions as well as lower pumping losses.

The trend of soot and NOx emissions agree with the results presented by Murata et al [210] for constant charge air pressure but the results presented here show no penalty of unburned emissions (HC and CO). This is believed to be a result of the fact that airflow was maintained and thus the air-fuel ratio was maintained. In conclusion, the results show that LIVC can also give both low soot and NOx emissions if the airflow and the CA50 are kept constant.

The effect of deactivating one intake valve and increasing the swirl number from 0.2 to 2.5 gave large increases in fuel consumption with the expected increase in NOx emissions for cases without EGR. There was no clear effect of increased mixing due to increased swirl on unburned emissions.

Figure 16. The effects of varied IVC on soot (filled symbols) and NOx (unfilled symbols) for 0 and 30% EGR at 50% load.

Figure 17. The effects of varied IVC on rate of heat release for 30% EGR at 50% load. The stars show CA50.

Figure 18. The effects of varied IVC on temperature for 30% EGR at 50% load.

Figure 19. The effects of varied IVC on CO emissions (filled symbols) and bsfc (unfilled symbols) at TDC for 0 and 30% EGR at 50% load.

Table 1. Measurement conditions. The engine speed was 1200 rpm and the injection pressure was 2400 bar. The charge air pressure was increased from a baseline of 1600 mbar (25% load) and 2400 mbar (50% load) to maintain the airflow.

Figure 15. The effects of varied IVC on pressure (filled symbols) and tem-perature (unfilled symbols) at TDC for 0 and 30% EGR at 50% load.

Conclusions and ongoing workThese results indicate that low temperature combustion using high levels of EGR is dif-ficult to achieve without significant HC and CO emissions and thus high fuel consumption. However, there is a possibility to operate the engine under partially premixed combustion at intermediate EGR levels where both NOx and soot emissions are fairly low and where CO and HC emissions have not yet begun to rise, resulting in low fuel consumption. The results also show that a combination of high NOP and intermediate EGR levels can reduce soot emis-sions without increases in NOx emissions.

In an effort to further reduce soot and NOx emis-sions, a measurement campaign was performed using the Cargine VVA system. It was found that when IVC is either retarded or advanced, the ignition delay is prolonged and the com-bustion becomes more premixed. For the cases without EGR a change of IVC had only minor


Diesel Engine Optimization

ObjectivesThe objective of this project is to develop a method to optimize the target set points of vari-ous controllable engine systems for a passenger car diesel engine. The optimization should be performed using a simulation model for a die-sel engine together with simulation models for the rest of a vehicles drive train. The engine calibration should be optimized such that the fuel consumption for a dynamic driving cycle is minimized, while regulatory requirements on emissions and other engineering requirements are fulfilled.

BackgroundModern passenger car diesel engines are grow-ing more complex in order to meet stricter regu-latory requirements on emissions and demand for lower fuel consumption. To respond to these additional requirements, more and more con-trollable systems are being added, leading to more degrees of freedom for engine operation. Examples of degrees of freedom in a typical passenger car diesel engine include controllable boost pressure, controllable exhaust gas recir-culation rate, controllable fuel rail pressure, and multiple injections with controllable timing and duration. These degrees of freedom provide the possibility to operate an engine more efficiently, but it also increases the complexity required

to optimize the calibration of the Engine Management System (EMS), i.e. to find control strategies and set points which lead to optimal operation of the combustion system.

To efficiently address this complex design task, model-based design strategies are a via-ble complement to more classical engine tests. Initial EMS design and optimization can be performed using simulation techniques for the relevant engine system in order to reduce the need for extensive engine tests. A simulation based approach can also be employed in early design phases, where different designs can be simulated and the results evaluated before building a physical engine. Using this approach there is a trade-off between model accuracy and simulation speed. The model accuracy needs to be good enough so that the optimized engine calibration in the simulation model is also valid for the real engine, but to be able to perform the optimization within a reasonable time; the simulation models also needs to execute as fast as possible.

Several types of engine simulations exist, ranging from detailed CFD calculation of the combustion process to purely data-driven map-based models. The very detailed models typi-cally suffer from long execution time, making

effects on in-cylinder pressure and temperature at TDC. However, when EGR was added, both early and late IVC give lower in-cylinder pres-sure and temperature. This is believed to be due to the lower airflow and thus lower charge air pressure for the EGR cases compared to the cases without EGR.

A change of IVC will reduce NOx emissions without bsfc penalties (except for very early or late IVC) and with increased exhaust tempera-tures for cases without EGR. Late IVC reduces soot emissions because a larger portion of the charge is burned under premixed conditions.

Thus late IVC combined with EGR can give reductions in soot with low NOx emissions (due to the EGR) with fuel consumption maintained and with increased exhaust temperatures.

For 25% load with 50% EGR, CO emissions were lower with EIVC compared to LIVC. The soot emissions were slightly higher than for cases without EGR but the NOx emissions were effectively zero. Furthermore, for these EIVC cases, there was no increase in fuel con-sumption, even compared to baseline without EGR. Thus, low engine out emissions can be achieved without increased fuel consumption for EIVC as well.

An early IVC appears to require higher charge air pressure to maintain the airflow constant through the engine for a similar, effective com-pression ratio. This results in increased pump-ing losses and higher fuel consumption for the early IVC. For early IVC, the expected extra cooling of the charge air during the expansion in the intake stroke appears to be counteracted by increased heat transfer from the walls. Thus, the temperature reduction at TDC is similar for early and late IVC.

Figure 20. The effects of varied IVC on soot (filled symbols) and NOx (unfilled symbols) for 0 and 50% EGR at 25% load.

Project leadersProf. Thomas McKelveyProf. Ingemar Denbratt (Project co-leader)

Industrial supervisorsKrister JohanssonJan-Ola Olsson

ResearcherMarcus Grahn (PhD Student)

Markus Grahn, Ph.D. student, Department of Signals and Systems/Volvo Car Corporation

them unsuitable for EMS optimization. The pure data-driven models are used for EMS opti-mization today [202,213], but are very sensitive to small changes in the air system, e.g. a resized turbocharger. Semi-empirical models for com-bustion have also been used for EMS optimiza-tion [214], however they have been limited to steady-state engine operation.

Project descriptionThe approach taken in this project is to imple-ment a complete simulation of a diesel engine powered vehicle system. The simulation model should be able to simulate a complete dynamic vehicle driving cycle while predicting fuel con-sumption, and NOx and soot emissions with input from the calibration of the EMS.

The simulation model will then be used to develop an optimization procedure for the engine calibration such that fuel consumption is minimized for a certain driving cycle, while accumulated emissions are kept below a given limit.

A schematic overview of the complete simu-lation model with its sub-models is shown in Figure 21.

Methods used for this yearThis year, the focus of the project has been to implement models for NOx and soot emissions, as the final stage of the modeling part of the project. The models need to estimate emissions accurately over a wide engine operating range. The models also need to be fast because they are intended to be used for engine calibration optimization.

The following existing models have been imple-mented and evaluated using engine measure-ment data that cover a wide range of the engine operating area.• A physical model for NOx emissions as

described in [217].• Semi-empirical models for NOx and soot

emissions as described in [218].• Global regression models for NOx and soot

emissions as described by [213].

Unfortunately, none of these models were able to well when evaluated using measured data from a Volvo passenger car diesel engine. Therefore, purely data-driven emissions models developed within VCC have been implemented.

Data-driven model for emissionsThe models for emissions that are needed for off-line calibration of EMS need to produce acceptable prediction accuracy for a wide area of the engine operation range, and preferably also be fast. One solution is to use data-driven modeling. Data-driven models are in general fast to execute, and are able to provide good prediction results for operating points close to operating points that have been used for calibra-tion of the models. The major downside with data-driven models is that a large amount of measurements have to be performed on a real engine to be able to calibrate the model over a wide area of the engine operating range.

Several data-driven models are described in the literature. However, detailed information regarding implementation of these models, the operating range of the engine that can be handled by the models, and model calibration is not available.

Here, a structure for data-driven models has been developed together with a method to calibrate them. The idea behind the model structure and calibration is that it should be straightforward to adapt the model to cover different degrees of freedom over the engine operating space, and that it should be suited for on-line implementation in the engine manage-ment system.

Model structureTo completely describe the operating range of combustion within a diesel engine, the operating area has to be defined using a multi-dimensional space spanned by all the variables that affect combustion. To completely cover the operating area of combustion by measurements is prac-tically impossible, since the amount of mea-surements needed becomes too large. However, several of the controllable input parameters that Figure 21. Schematic overview of the

complete simulation model with its different sub-models.


affect emissions are usually calibrated based on the engine speed and injected fuel work-ing point. This means that for an engine with a given EMS setting that is only operating on steady-state points, a single two-dimensional grid-map of engine speed and injected fuel as inputs would be sufficient to estimate emissions.When an engine is operated under transient conditions, a simple map is no longer suffi-cient to estimate emissions. During a transient in the speed and fuel space, the combustion conditions are not the same as the steady-state conditions owing to the dynamics of the engine

air system. If a model is meant to be used for off-line engine calibration, it also needs to react on the controllable engine parameters that should be optimized. This means that to be able to use the model for calibration of controllable engine parameters and to be able to use the model for transient engine behavior, a simple interpolation map in the speed and fuel domain of the engine is not sufficient.

Since the control structure of an engine is usu-ally based on the engine speed and the injected fuel amount, the structure of the model for emis-sions should be based on the same principle. For a certain engine speed/injected fuel work-ing point of the engine we assume that one can model the emissions with a linear regression using deviations from nominal values for input signals that affect the emissions.

(1)

where y denote the predicted emission, the various βi are model constants and the zi are deviations from the nominal values for input signals that affect the emissions.

We now assume that the nominal emissions change depending on the speed/fuel point of the engine and the deviations in the input sig-nals, i.e. that the parameters in the regression model are different for different speed/fuel points of the engine. Therefore we choose to define the regression parameters as grid-maps in the speed/fuel domain of the engine. We get

(2)

where x1 and x2 are the input signals for engine speed and injected fuel respectively, zi are the deviations from the nominal values for the other input signals to the model, and βi are the regression parameters represented by two-dimensional grid-maps. The deviation signals, zi, could be chosen to represent either deviations of different properties due to transient engine operation or deviations due to different calibra-tion of engine parameters.

This means that we have a grid-map for emis-sions together with additive correction terms for signals that deviate from their nominal val-ues. The influences of the various deviations are dependent on the speed/fuel working point for the engine. The reason for choosing this structure is that all the parameters that affect emissions but are not included in the base model usually are calibrated based on the speed/fuel of the engine. Parameters that vary throughout

Figure 22. Measured NOx mass flow versus estimated NOx mass flow from the global model for NOx emissions.

Figure 23. Measured soot mass flow versus estimated soot mass flow from the global model for soot emissions.

the engine operation, but depend only on the engine speed and injected fuel, do not have to be directly considered in the model.

The model structure presented here also leads to grid-maps that have meaningful interpre-tations. The grid-map β0 can be interpreted directly as the modeled emissions when the engine is operated in steady-state at nominal conditions, and the grid-maps βi for i>0 can be interpreted directly as how the deviation from nominal values for various input signals affects the emissions.

ResultsThe performance of the data-driven models for NOx and soot emissions has produced better estimation than any of the other models that have been evaluated. The modeling results for the data-driven models for NOx and soot emis-sions are illustrated in Figure 22 and Figure 23 respectively.

Conclusions and ongoing workThe modeling part of this project has been final-ized. A complete simulation model for the diesel engine vehicle system has been implemented.Work has now started with the second part of the project, i.e. developing optimization procedure for the engine calibration.

Project leaderProf. Ingemar Denbratt

ResearcherMonica Johansson (PhD Student)

Alternative Fuels

Monica Johansson, Ph.D Student,Division of Combustion, Department of Applied Mechanics, Chalmers University of Technology.

Objectives The objective of this study is to investigate the influence of neat and low blends of biofuels in a Diesel engine and compare the resulting emissions and combustion characteristics with conventional Diesel fuel (EN590). The project is a collaboration between three research projects in CERC, involving simulations of both spray chamber and single cylinder engine, spray stud-ies in spray chamber with advanced laser diag-nostics, and combustion and emissions studies in heavy-duty engine.

Background The European directive from 2009 specified that by the year 2020 the transport sector should use a minimum of 10% renewable energy. Biofuels are critical to attaining this target. Combustion of oxygenated biofuels such as FAME (fatty acid methyl ester) also produces less soot but it can also increase NOx emissions.

Prior literature demonstrated that the engine operating mode and type of injection system are important factors for the increase or decrease of NOx emissions with an increase in FAME concentration. For conventional pump-line-nozzle fuel injection systems the bulk modulus of compressibility of the fuel impacts injection timing. A higher bulk modulus advances the injection [222]. The advance in fuel injection timing then increases the NOx emissions. For common rail injection systems, however, the bulk modulus does not affect the start of injec-tion [223]. Mueller concluded [221] that since oxygenated fuels produce less in-cylinder soot the flame does not radiate as much energy and hence the flame temperature is higher, leading to higher levels of NOx in comparison to nor-mal diesel fuel combustion. Moreover Mueller found that in the premixed autoignition zone,

near the flame lift-off length, the reacting mix-tures are closer to stoichiometric for biodiesel fuels. This would lead to higher local and aver-age in-cylinder temperatures and hence higher NOx emissions in comparison to diesel fuel.

Increasing the biodiesel blend ratios into con-ventional diesel fuel often produces an increase in engine out NOx emissions [223]. There are studies, however, where the increase is not sig-nificant. The ambient temperature, for example, has been shown to be an important factor; for example at -5°C the increase in NOx emissions is not significant [219]. A report from the National Renewable Energy Laboratory [220] investi-gated different vehicles and driving cycles for B20 fuels and there was no significant indica-tion that Biodiesel increases the NOx emissions.

The alternative fuel project at CERC has the intention to investigate low blends of FAME fuels in European diesel (EN590) used in a heavy-duty single cylinder engine. In order to explain combustion behavior and emis-sions formation for the different blends, spray experiments in Chalmers’ high temperature and high pressure chamber were performed (Raul Ochoterena). Moreover CFD simulations were used to enhance the understanding of emission formation for different blends of biodiesel in conventional diesel (Junfeng Yang).

Method Neat FAME fuels (RME, PME and SME) and RME blends (7% and 30%) with European die-sel (EN590) were tested in a heavy-duty single cylinder engine. Specification of the FAME fuels and the Diesel fuel can be seen in Table 1.The engine used in the experiments was an AVL 501 single cylinder engine with a displacement


Table 1. Fuel specifications.

of 2 liters. The engine was equipped with a D12C 4-valve cylinder head from Volvo Powertrain. The compression ratio was 17:1. The injection system was an electronically controlled unit injector with a six-orifice nozzle.

The engine operational points were chosen from the European Stationary Cycle (ESC). In Figure 24 all the measuring points in the cycle are shown as green circles and the ones marked red circles were chosen for this study; A25, B50, B75 and C100.

The A25 case was also extended to allow varia-tions in the start of injection (SOI), the NOx emissions, and the needle opening pressure (NOP) according to Figure 25.

Figure 24. European Standard Cycle operational cases.

Figure 25. Design of experiments for extended A25.

Results Soot emissions for the various load cases are shown in Figure 26 for diesel, RME, SME and PME fuels. As shown, all of the FAME fuels reduce soot emissions, in all of the load cases. In contrast to FAME fuels, diesel contains aro-matics, which are precursors to soot in flames. Since the FAME fuels do not contain aromatics they produce less soot. The presence of oxy-gen in the FAME fuel makes the equivalence ratio of the spray leaner and the leaner spray also enhances combustion producing lower soot emissions in comparison to diesel fuel. In relation to diesel, RME, SME and PME lower the soot emissions by more than 60, 50 and 70% respectively. Moreover, the higher density and higher boiling intervals of the FAME fuels lengthen the liquid penetration, allowing more air entrainment into the spray before evapo-ration, further decreasing soot production. It was also shown that low blends of RME with Diesel (7 and 30%) reduced soot emissions significantly.

The CFD simulation in Figure 30 shows the soot concentration in the piston bowl at three crank angle positions (1.5, 5 and 10 CAD ATDC) for Diesel and RME100 at the B50 operational case. Red indicates high soot concentration while blue indicates low. The figure shows that the RME fuel produces less soot than the Diesel fuel does. Faster penetration for RME can also be observed.

The flame lift-off lengths measured in the spray chamber for Diesel, RME7 and RME100 fuels are shown in Figure 28. Lift-off is defined here as the distance from the injector orifice exit to the flame stabilization point. Longer lift-off generates increased air entrainment and hence lower soot emissions. It shows that only a small fraction of RME (7%) increases the lift-off length and air-entrainment, reducing soot production and (via engine experiments) the engine out soot-emissions.

NOx emissions for the FAME fuels, for all the operational cases, are shown in Figure 29. At point B50, the figure indicates that all of the FAME fuels generate higher (12-18%) engine-out NOx emissions in comparison to diesel fuel. However, in the B75 load case it shows that the PME fuels generate lower NOx while the RME fuels generate higher NOx emissions than the Diesel fuel. The explanation for the different trends at the different load cases can be based on combustion speed. In the B50 load cases, all FAME fuels burns faster than Diesel, while at B75 operation point PME and Diesel have similar combustion speed.

Figure 27. Simulation of the soot concentration in the piston bowl for diesel and RME100 at the B50 operational case.

Figure 28. Flame lift-off length for Diesel, RME7 and RME100.

Figure 26. Soot emissions for all of the load cases.

The B50 operational case was further simu-lated using CFD with simulations of the equiva-lence ratio (φ) for Diesel and RME fuel shown in Figure 30. Red symbolizes high φ and blue symbolizes low φ. As shown, the RME fuel produced lower equivalence ratios (φ, i.e. leaner local reacting mixture) than the Diesel fuel and hence this can explain why more NOx is formed with RME combustion than with diesel.

From the spray-rig experiments it can further be concluded that the RME fuel has a higher flame temperature (measured via 2 wavelength pyrometry) than Diesel. Figure 31 illustrates the average flame temperature for Diesel, RME7 and RME100. The flame temperature is approx-imately 50 degrees higher for RME than Diesel, which can also help explain the increased NOx emissions for RME.

In Figure 32 the NOx emissions for the RME fuels are illustrated. As shown the NOx emis-sions in the B50 load case for RME7 are lower than the diesel case, and for RME30 they are equal to the diesel NOx. In the C100 case, the NOx emissions are similar among all of the fuels which indicates that a total ESC test (or even better, a transient cycle) would be needed to fully conclude if RME7 lowers the NOx emis-sions in relation to diesel.

Brake specific fuel consumption values for the FAME and diesel fuels are shown in Figure 33. Fuel consumption increased for all of the FAME fuels by 12-18%, and when the heating value of the fuels (13% lower for the FAME fuels) is taken into account, the fuel consumption is even higher than Diesel (except for load case C100). The explanation for this trend can be that the fuel economy is not optimized for the FAME fuels at these standard settings for the load cases as it is for diesel.

Figure 31. Average flame temperature for Diesel, RME7 and RME100.

Figure 30. In-cylinder equivalence ratio distributions for the Volvo D12C Diesel engine fuelled by a) Diesel, b) RME at -1.5 CAD ATDC, under engine operating condition B50 (Junfeng Yang).

Figure 29 (left). NOx emissions for all the operational cases.


Conclusions The conclusions from the last year can be sum-marized in the following points:• RME, SME and PME fuels lower the soot

emissions by more than 60%, 50% and 70% respectively, in relation to conventional die-sel, in the B50 case. Moreover in the C100 case the soot emissions are reduced up to 90% for the FAME fuels. Even small blend ratios of RME in diesel show significant reduction of soot emissions.

• The explanation for reduced soot emissions for the FAME fuels is their oxygen content, which enhances the oxidation of soot precur-sors and soot particles.

• The longer flame lift-off of the FAME fuels is also a part of the explanation for the low soot emissions since it enhances the entrain-ment of air into the spray.

• The NOx emissions increase by 10-20% for all neat FAME fuels compared to diesel. The increase in NOx emissions for the FAME fuels is mainly due to the oxygen content in the fuels leading to a lower equivalence ratio (φ), i.e. leaner local reacting mixture, than the Diesel fuel which leads to more NOx formed and higher combustion temperature (50°C).

• Fuel consumption increased for all of the FAME fuels, the increase is due to the lower heating value but also the fact that the set-tings and engine are not optimized for FAME fuels.

Figure 33. Brake specific fuel consumption at all load cases for the FAME fuels and diesel.

Figure 32. NOx emissions for RME blends for operational cases B50, B75 and C100.

Spray Turbulence Interaction

Project leadersProf. Ingemar DenbrattAdjunct Prof. Anders Karlsson (Project co-leader)

ResearcherAnne Kösters (PhD Student)

ObjectivesThe main focus of this project is to address the shortcomings of current models in order to better predict Diesel engine fuel efficiency and emissions by CFD.

BackgroundSpray formation and combustion are highly complex processes and the modeling of these processes is a challenging task. Modeling of Diesel combustion is one of the key areas within CERC; models for various phenomena have been implemented in multi-dimensional CFD-codes like e.g. KIVA or the open source code OpenFOAM [224,225,80]. Several prob-lems remain, however, and the goal of this proj-ect is to investigate the spray-turbulence and turbulence-chemistry interaction, using the open-source software package OpenFOAM®

[226]. Turbulence enhances mixing of fuel and air, and strongly influences the combus-tion process. A good understanding of these processes is needed to accurately describe them. That point motivates a closer consideration of the spray-turbulence and turbulence-chemistry interactions. Appropriate CFD models that can describe the interaction and hence enable fur-ther investigations are needed. The first part of this project involved the implementation of the VSB2 spray model and results of that work are published in [151]. The model was chosen based on its robustness and minimal number of tuning parameters. Moreover, grid dependence was reduced. It has already been shown in [81] that the model is able to describe spray forma-tion in engine simulations and can be applied over a wide range of conditions.

The obvious next step was modeling of turbu-lent combustion. It is very challenging since the turbulent mixing of vapor fuel with air and the interaction of the turbulence with the chem-istry must be described. There exist different approaches to modeling turbulent combustion like the probability density function (pdf), the flamelet model, and well-stirred and partially stirred reactor models. In [227] for example, Kokjohn and Reitz use a well-stirred reactor approach without considering subgrid turbu-lence-chemistry interaction. Hong et al. [228] apply a partially stirred reactor model based on the eddy dissipation concept (EDC) [229]. The EDC is a known technique to treat turbulence-chemistry interaction with the inclusion of sub-grid scale turbulence-chemistry interaction.

The results published in references [224,225,80] are all based on the EDC . The models define a reactor volume based on chemical and mixing time-scales, which are very difficult to deter-mine. The Volume Reactor Fraction Model (VRFM) was chosen in this work; because it defines the reactor volume based on the prod-ucts of the complements of unmixedness in mixture space and in progress space. With this definition, the problem of finding representative time-scales for mixing and chemical reaction is removed. Furthermore, the approach correctly accounts for the unmixed situation created by evaporation, a feature that the traditional time-scale expressions for the reactor volume can-not describe. Results obtained with the VRFM model were compared to experiments carried out at Sandia National Laboratories within the engine combustion network (ECN) [230]. Flame lift-off was compared for different EGR levels and at different ambient conditions. Flame lift-off and stabilization are of particular interest in the case of the turbulent combustion process occurring in a diesel engine. The lift-off length is an indicator for how well the fuel and air are mixed before the mixture reaches the flame. When fuel is well mixed with air, the soot for-mation during combustion is reduced.

MethodsThe Volume Reactor Fraction Model (VRFM) has been implemented in OpenFOAM® [151]. The current model was developed by Karlsson as an extension of the model he presented in [224]. The model is based on the Eddy Dissipation Concept (EDC) by Magnussen [229], but con-siders the non-uniformities of the fluid cell. The main idea of the model is to define a reactor with a volume that is smaller or equal to the volume of the grid cell as shown in Figure 34. The main difference between the current VRFM model and the model presented in [224] is the defini-tion of the reactor volume.

The size of the reactor volume is defined using к (see Figure 34), as shown in following equation:

(1)

whereZ̃ is the mixture fraction, is the variance of the mixture fraction,c̃ is the chemical progress variable, and is its variance.

The mixture fraction can be understood to be the local ratio of injected fuel mass to the entire mass. The chemical progress variable provides information about the chemical state. Hence both mixing and chemistry are considered. The reactor properties (e.g. the mass fractions in the reactor, YR, cp. Figure 35) are then used as the input to the chemistry calculation. The chemistry calculation gives the rate of change of the species concentration within the reactor, , which is then included as a source term in the transport equation of the reactor species concentration, YR, (cp. Equation (2)). Chemical reactions are simply progressing in the reactor, and via turbulence the reactor properties mix with the mean properties within the cell, cp. Figure 35 and equation (2).

The transport equation of the mean species con-centration is given by equation (3). The chemi-cal source term ωωi is included considering the reactor volume with к (equation (3)).

(2)

(3)

Figure 34. Definition of a reactor volume.Figure 35. Definition of reactor properties, e.g. the species mass fraction in the reactor, YR.

Anne Kösters, PhD Student,Division of Combustion, Department of Applied Mechanics, Chalmers University of Technology.


A reduced chemical mechanism of n-hep-tane, C7H16, was used in the simulations. The mechanism was reduced by Prof. Golovitchev and contains 36 species and 74 reactions. The mechanism was validated against experimental data for auto-ignition and flame propagation, as shown in Figure 36.

ResultsThe VRFM combustion model was validated by comparison to flame lift-off experiments done at Sandia National Laboratories within the Engine Combustion Network [227]. Flame lift-off of n-heptane sprays was measured under varying ambient temperature, pressure, and EGR levels. The nozzle orifice was 100 μm and the injection pressure was around 1500 bar. Tables 1 and 2 show the experimental conditions.

Mole fraction O2 [%] Tgas [K] pgas [bar] pinj [bar]

21 967 42.1 1502

15 967 42.5 1533

12 967 42.7 1533

10 967 42.8 1533

Mole fraction O2 [%] Tgas [K] pgas [bar] pinj [bar]

15 962 87 1557

12 962 87.4 1557

10 962 87.6 1568

8 962 87.9 1558

Figure 37 shows the results of the lift-off length obtained via simulation compared to the experi-mental results. At both ambient pressures the lift-off length was predicted well for various EGR levels. Flame lift-off decreases with increasing mole fraction of O2, which corre-spond to decreasing EGR.

Figure 37. Flame lift-off at different conditions, comparison between experiments and simulations.

Figure 38. Different fields at 3.5 ms after start of injection, 15% O2 in air, Tgas=962K, pgas=87bar.

Table 1. Experimental conditions, pgas=42 bar.

Table 2. Experimental conditions, pgas=87 bar.

Figure 36. Comparison of calculations based on the reduced n-heptane oxidation mechanism with experimental results.a) Ignition delay for autoignition of PRF0/air mixtures in Gauthier shock tube experiments at 15–20 bar and 45–60 bar with calculated results at 20 bar and 52 bar.b) Flame propagation speed in n-heptane/air mixtures (P0 = 1 bar, T0 = 298 K).

Figure 38 shows the temperature field, the OH concentration, values of к (kappa), the mixture fraction Z̃ , and the chemical progress variable c̃ at 3.5 ms after start of injection and with 15% O2 in the air. The OH field indicates stoichio-metric mixtures. The high OH concentration at the flame-lift off position is consistent with the flame stabilization point. At the stabiliza-tion point combustion occurs under premixed conditions, which explains the peak in OH for-mation. Kappa is dependent upon the mixture fraction Z̃ , and the chemical progress variable c̃. At the outside it is mainly controlled by the chemical progress and in the core of the spray by the mixture fraction.

ConclusionsThe VRFM combustion model was imple-mented in OpenFOAM and n-heptane com-bustion simulations were performed under varying ambient conditions. Results of flame lift-off length simulations were compared to experiments carried out at Sandia National Laboratories. The results of the simulations agree well with the experiments. The effect of EGR on lift-off length was captured in the simulations. The predictions of к, the mixture fraction Z̃ , and the chemical progress variable c̃, are reasonable.

Ongoing workAfter the implementation of the VSB2 spray model and the VRFM combustion model, the next step is the implementation of the represen-tative interactive flamelet (RIF) model. After the implementation of the RIF model, a visit to RWTH Aachen is planned to further work with the RIF model. A licenciate seminar is also planned in February, 2012.

a) b)

Project leaderAssoc. Prof. Valeri Golovitchev

ResearcherJunfeng Yang (PhD Student)

Combustion Models for Bio-Diesel Fuels

Junfeng Yan, Ph.D. student, Division of Combustion, Department of Applied Mechanics, Chalmers University of Technology.

ObjectivesThe primary purpose of the present work was to derive a skeletal RME combustion mecha-nism from a detailed master biodiesel model and to validate it by comparing predicted ignition delay times and in-cylinder parameters under typical diesel engine operating conditions. Based on this semi-detailed combustion mecha-nism, better insight of importance to physical spray dynamics and combustion chemistry can be achieved through 3D CFD simulations in a constant-volume IQT combustion chamber

BackgroundIgnition delay is crucial factor affecting the in-cylinder combustion and emissions formation processes in diesel engines. It comprises both physical (e.g., spray breakup, droplet vapor-ization, air entrainment) and chemical kinet-ics processes. These processes are normally coupled, making it impractical to identify the relative influence of each on overall ignition delay via experiments. To overcome this chal-lenge, 3D CFD modeling and 0D parametric studies need to be performed; to handle the problem of estimating characteristic times for droplet formation, droplet evolution, and chemistry. The overall ignition delay can be defined through a constant-volume combustion model of the Ignition Quality Tester (IQT), a device used to estimate the cetane number (CN) of diesel-like fuels. The droplet evolution and chemical reaction times can be estimated by investigating single droplet breakup, life time, and chemical kinetics studies. Such research for n-heptane has been reported in [231]. However, no relevant research has been carried out for biodiesel fuels.

An accurate combustion mechanism is required to model biodiesel spray combustion process in the IQT. Recently, a master combustion mecha-nism including 5-components (methyl palmi-tate, C17H34O2, methyl oleate, C19H36O2, methyl linoleate, C19H34O2, methyl stearate, C19H38O2 and methyl linolenate, C19H32O2) representing RME and SME (if taken in proper proportions) has been constructed by Westbrook et al. [232]. However, the large number of species involved in this detailed model makes its implementation extremely time-consuming in 3D CFD simula-tions. Therefore, a mechanism reduction strat-egy is required to simplify the mechanism to a reasonable size, ideally maintaining similar combustion features in comparison to the mas-ter mechanism, e.g. ignition delay behavior and PSR species profiles. Mechanism validation by calculating the premixed laminar flame velocity and 3D engine simulations are also necessary to ensure an accurate combustion model before IQT simulations.

MethodsFirst, IQT modeling based on the original master mechanism for biodiesel combustion predicted very low CN [233]. Therefore, the master mech-anism was adjusted as follows: the rate constants (A-factors) for RO2 (radicals of hydroperoxide methyl esters) and O2QOOH (ketohydroperox-ide) isomerization were increased by a factor of 15, making the O2QOOH heat of formation well deeper by 4 kcal/mol compared to original values, whereas the rate constants including ROOH (hydroperoxide methyl esters) and keto-hydroperoxide were increased by 10 times. In addition, the rate constants (A-factors) of the most sensitive reactions in the high-temperature


sub-mechanism were increased by a factor of 2. The final tuned RME mechanism yielded CN that was close to that for real RME [234]. Meanwhile, these adjustments did not affect the fuel conversion rate for the temperature range 650 K – 1200 K measured in a jet-stirred reactor.

Second, methyl oleate (C19H36O2), molecular structure see in Figure 39a, was selected as a surrogate fuel for RME, and the mechanism reduction procedure was applied using the Reaction Workbench [235] at closed homoge-neous reactor conditions similar to that used in shock-tube ignition delay modeling. Parameter values were selected to represent typical con-ditions in diesel engines prior to combustion, i.e., ranges of temperature (600-1200 K) and equivalence ratio (0.5-3) at a pressure of 50 bar. The reduction strategy methods employed in the present work include the directed relation graph (DRG) method, the enhanced DRG method called DRGEP (directed relational graph with error propagation), and isomer lumping. The smallest reduced mechanism for methyl ole-ate (found via CHEMKIN-Pro [240]) involved only 491 species (over 90% reduction) with an accurate prediction of auto-ignition delay times (see Figure 39b) and fuel conversion rates (see Figure 40).

Third, the reduced methyl oleate mechanism was validated further against experimental data for the Volvo D12C heavy-duty diesel engine fuelled by RME. The 3D CFD diesel engine modeling and IQT simulation mentioned below were carried out using the FORTÉ software package [241]. The engine modeling results were presented in Figures 41-42 and Table 1.

Forth, IQT modeling can provide overall igni-tion delay times, but it can also help evaluate the accuracy of overall model involved: the breakup, evaporation and combustion sub-models for biodiesel fuels. The standard setting of IQT (e.g. air temperature and pressure) is specified according to the American Society for Testing and Materials (ASTM) method D6890-11 [242]. Typical combustion pressure and needle-lift traces used to determine the ignition delay are shown in Figure 43a. In the present study, the ignition delays were defined as the time from the start of injection (SOI) to the combustion pres-sure recovery point, ~2.4 bar. The correlation of predicted ignition delay times with Derived Cetane Number (DCN) is given in Figure 43b according to empirical relationships defined by ASTM D6890-11 [242]. The IQT modeling presents an overall ignition delay prior to the main combustion event. The 0D auto-ignition parametric study has been implemented on a closed homogeneous reactor [240] to obtain the pure chemical reaction times for RME fuel. The droplet breakup and evaporation times will be studied in future work.

ResultsFigures 39b and 40 show comparisons of the predicted pure chemical ignition delays and fractional conversion of methyl oleate using the reduced and the tuned master mechanisms under the same conditions. The results show that the reduced mechanism performs well compared to the tuned master mechanism over the entire range of conditions considered. Furthermore, this reduced biodiesel mechanism was validated under the diesel engine operating conditions in Volvo D12C engine. The predicted in-cylinder parameters (pressure and rate of heat release) are in a close agreement with measured data [42], see Figure 41. Table 1 presents the cal-culated emission concentrations compared to measured values [42]. Soot emissions were only predicted approximately due to limitations of the soot formation/oxidation model used. NOx emission was predicted reasonably well for both diesel oil surrogate (fuel for a comparative study) and RME surrogate. For other emissions, CO and UHC, the RME mechanism over-predicts CO concentration by nearly 8 times and under-predicts UHC concentration by a factor of 2. The 3-component diesel oil mechanism [236] predicted CO concentration fairly accurately, but it also under-predicted UHC emission. The

in-cylinder temperature contours and droplet breakup for RME and diesel oil combustion are presented in Figure 42 for comparison.

Figure 44 presents the IQT modeling results for methyl oleate at time instant 3.7 ms ASOI which is identified as the combustion pressure recov-ery point. Hence, the ignition delay time of 3.7 ms is reported here for methyl oleate. According to Figure 43b, a DCN value 54.9 for methyl oleate was predicted, in fairly good agreement with the measured value (56) reported in [234]. Moreover, Figure 44a illustrates that the equiva-lence ratio of the reacting mixture covers the range (1~3) prior to ignition. Combustion hap-pens at the tip of the spray, see Figure 44b-c. The 0D auto-ignition parametric study demon-strates that the pure chemical ignition delays of methyl oleate in the relevant equivalence ratio range vary from 0.95 ms at ϕ = 1 to 0.52 ms

Figure 39. The a) ball and stick mode structure of methyl oleate, C19H36O2, red color represent oxygen atoms and bonds, blue color - carbon atoms and bonds , white color - hydrogen atoms and bonds; the valency electrons made visible; b) auto-ignition delay times for a stoichiometric methyl oleate/air and reference fuel mixtures. The original 5037 species mecha-nism is represented by the bold line with cross symbols [232], the tuned 5037 species mechanism by the bold line, and the 491 spe-cies mechanism by the dashed line).The auto-ignition delay times for a diesel oil/air mixture (437 species mechanism) under 13.5 atm is shown by thin line [236]. Also shown are experimental data for stoichiometric n-heptane/air (circles) [237] and n-decane/air (triangles) [238] mixtures.

Figure 41. Volvo D12C diesel engine modeling results: comparison of cal-culated and measured parameters.a) In-cylinder pressure vs. CADs.b) RoHR vs. CADs.

Table 1. Volvo D12C diesel engine modeling results: comparison of cal-culated and measured in-cylinder emission concentrations: soot, NOx, CO and UHC, for a 25% load, 25% EGR level and SOI=-3.3 CAD ATDC.

Figure 42. Comparison of predicted in-cylinder temperature and droplet distribution for Volvo D12C diesel engine fuelled by RME: a, c, e and g); by diesel oil, b, d, f and g) under 25% load, 25% EGR level.

Soot, g/kw-h NOx, g/kw-h CO, g/kw-h UHC, g/kw-h

Exprt. Calc. Exprt. Calc. Exprt. Calc. Exprt. Calc.

RME 0.00397 0.0229 3.164 3.301 0.57 3.88 0.0127 0.0078

Diesel oil 0.0229 0.0202 2.441 3.084 0.78 1.072 0.0395 0.027

b)

a)

Figure 40. The fractional conversion of methyl oleate in a jet-stirred reaction. Lines are computation-al results for original 5037 species mechanism, bold line with cross symbols, tuned 5037 species, bold line, and 491 species mechanism, dashed line. Triangles show measured data from [239] for methyl oleate. The total mole fraction of fuel (0.26 methyl oleate/0.74 n-decane) was 0.002 in the stoichiometric O2/He mixture at atmospheric pressure [239].

a) b)


Project leaderProf. Ingemar DenbrattAssoc. Prof. Andrei Lipatnikov(Project co-leader)

ResearcherChen Huang (PhD Student)

Modeling of Gasoline Direct Injection Spark Ignition Engines

ObjectivesThe goals of the project are 1. to develop models and methods that can be

used for simulations of direct injection (DI) spark ignition (SI) engines that burn various fuels, e.g. gasoline, ethanol, and their blends,

2. to implement these models into publically available CFD library called OpenFOAM,

3. to apply the extended code for unsteady 3D RANS simulations of fuel injection, turbu-lent mixing, and combustion in a DI SI engine and to compare the computed results with experimental data obtained within the framework of another CERC project entitled “Spray-Guided Gasoline Direct Injection”.

BackgroundDirect injection of a fuel into the combus-tion chamber of a SI engine is considered to be a promising technological solution aimed at decreasing fuel consumption and hydrocar-bon emissions, increasing fuel economy, and improving efficiency owing to (i) the reduc-tion of pumping losses by removing the throttle, (ii) cooling of charge due to the evaporation of fuel spray, (iii) higher compression ratios are allowed by the reduction of knock propensity, (iv) the elimination of over-fuelling during cold start, etc. [243]. Among several DI technologies for SI engines, spray-guided (SG) direct-injec-tion is currently considered to have the highest potential to increase fuel economy, as this solu-tion offers an opportunity to accelerate burn-ing, to reduce cyclic variability, and to lower unburned hydrocarbon emissions by formation of a compact fluid cloud around the spark. For instance, fuel economy can be improved by 20% in SG DI engines when compared to a throttled, port-fuel-injection engine on the New European Driving Cycle [244].

To realize the high potential of the SG DI SI technology, the automotive industry has a strong need for advanced CFD tools that numerically investigate fuel injection and evaporation, tur-bulent mixing, turbulent burning of stratified charge, and pollutant formation in the combus-tion chamber. For these purposes, both powerful CFD codes and advanced numerical models should be developed. This need is addressed by the present project focusing on (i) applying the open source code OpenFOAM to unsteady mul-tidimensional RANS simulations of mixture formation and combustion in a DI SI engine and (ii) developing and validating advanced models of turbulent combustion of stratified mixtures (including emissions).

Although several mature commercial CFD codes have been developed and are widely used for simulations of SI engines, there is a

need for less expensive software. OpenFOAM released in 2004 and free available online [226] is capable for satisfying such a need. The code has already been successfully applied to multi-dimensional numerical simulations of combus-tion in Diesel engines, but it should be substan-tially improved in order to become a R&D tool in the SI engine branch of automotive industry. To the best of our knowledge, OpenFOAM has not yet been used to model turbulent combustion in a gasoline DI engine. Such improvements involve development and implementation of predictive models of turbulent partially-pre-mixed flames, development and validation of a tractable semi-detailed chemical mechanism of combustion of gasoline/ethanol/air mixtures, accurate implementation of spray models used by automotive industry, etc.

In particular, several multidimensional RANS simulations for turbulent combustion of strati-fied mixtures in a SI engine [245,164] have been used by various research groups, but these models require quantitative validation against experimental data obtained in well-defined, simplified cases. In contrast, the capabilities of the so-called Flame Speed Closure (FSC) model developed at Chalmers for quantitatively predicting important features of stratified com-bustion in a SI engine have been thoroughly validated against a wide set of experimental data obtained by different research groups from laboratory confined expanding flames [164,246]. This includes dependence of turbu-lent burning velocity on mixture composition, pressure, and temperature, the development of a turbulent flame kernel after spark ignition, and so forth. This advantage of the FSC model makes it particularly interesting for simulat-ing stratified combustion in a SG DI SI engine. However, before doing so, the model should be extended to address complex combustion chemistry and emissions from turbulent flames. The present project is aimed at filling this gap.

Methods During the first year of the project (2009), a semi-detailed chemical mechanism of combus-tion of gasoline/ethanol/air mixtures was devel-oped and thoroughly validated against a wide set of experimental data on ignition delay times and laminar flame speeds, reported by several research groups. The results of this study are summarized in [128, 165].

During the second year of the project (2010), CFD work was focused on applying the OpenFOAM code to simulations of hollow-cone gasoline and ethanol sprays discharged by a piezo-actuated outward-opening pintle injector, which had

Chen Huang, Ph.D. Student, Division of Combustion, Department of Applied Mechanics, Chalmers University of Technology.

at ϕ = 3. It shows the chemical ignition delay constitutes less than 20% of the overall ignition delay time in this case. Since the 0D study of a droplet dynamics based on the d2 vs. time law has not been carried out for biodiesel fuel, a definite conclusion on a relative influence of chemical auto-ignition and droplet evolution needs further consideration.

Figure 43. The Ignition Quality Tester parameters: a) typical combustion pressure and needle-lift traces used to determine ignition delay, b) ASTM D6890 empirical correlation for DCN vs IQT ignition delay times.

Conclusions and ongoing work1. The original biodiesel mechanism (5037 spe-

cies/19990 reactions) was found to yield an unrealistically low CN based on initial engine tests and IQT simulations. To increase the apparent oxidation rate and simulate rea-sonable IDTs, a few reaction rate constants in the mechanism were optimized for low and high temperature conditions. The result-ing tuned biodiesel mechanism gave much shorter IDTs at low temperatures than the original mechanism. Jet-stirred reactor mod-eling produced good agreement with the data and demonstrated that the modifications made in the tuned mechanism had a negli-gible effect on the fuel conversion rate, which had been used to validate the original model.

2. The optimized biodiesel mechanism was reduced to a 491-species mechanism using a combination of reduction methods. This 491-species mechanism retained the same IDT characteristics and fuel conversion rates as the detailed mechanism of methyl oleate combustion. This reduced RME surrogate mechanism was applied to a Volvo D12C engine operating under typical conditions. Overall the results indicate that the reduced RME mechanism is useful for efficient and accurate engine simulations, and the model may be improved further by paying particu-lar attention to the sub-mechanisms involv-ing low-level species.

3. Based on this semi-detailed combustion mechanism, IQT modeling for methyl oleate was performed. The model predicts that the combustion event is governed by auto-igni-tion and droplet evolution (breakup and evaporation) events. The chemical ignition delay time constitutes less than 20% of over-all ignition delay times in this case. Note that, the predicted chemical ignition delays are highly dependent on the accuracy of the chemical model. A comparative study of different biodiesel models needs to be performed.

AcknowledgementThis project is partially funded through the VINNPRO program of the VINNOVA (Swedish Governmental Agency for Innovation Systems). The mechanism’s adjustment, reduction and engine validation work has been carried out through a 3-month visiting research at Reaction Design Inc. The authors also thank Dr. C.V. Naik from Reaction Design Inc., for sharing the case file for IQT simulations. Ms. M. Johansson was acknowledged for sharing the experimental results (in-cylinder pressure, RoHR and emis-sions concentration) for Volvo D12C Diesel engine.

a) b)

a)

b)

c)

Figure 44. The IQT modeling re-sults for methyl oleate, a) T vs. Equivalence ratio, b) temperature contours, c) equivalence ratio con-tour, at time instant 3.7 ms ASOI (after start of injection), each red cross symbol corresponds a com-putational cell.


been investigated experimentally by Hemdal et al. [96] in Chalmers spray chamber within the framework of another CERC project entitled “Spray-Guided Gasoline Direct Injection”. The numerical results are summarized in [165-168].Moreover, in order to extend the Chalmers model of premixed turbulent combustion, two alternative approaches, conditioned balance equations and a simple algebraic model, to evaluating turbulent scalar flux in premixed flames were developed and validated in [169-171] and [172], respectively.

In the first half of 2011, simulations of the afore-mentioned spray experiments were completed, papers [166-168] were submitted, and a licenti-ate thesis was written by Chen Huang [165].

In the second half of 2011, the CFD work was mainly focused on implementing the Chalmers model of premixed turbulent combustion [246] into OpenFOAM, improving implementation of other combustion-related libraries of the code, and debugging the extended code.

In addition, the compatibility of the aforemen-tioned simple algebraic model of turbulent sca-lar flux [172] with Chalmers model of premixed turbulent combustion was investigated and the former model was improved and validated [173,174].

The available presumed probability density function (PDF) approaches to modeling pre-mixed, partially premixed, and non-premixed turbulent combustion have also been assessed based on the following reasoning. In contem-porary simulations, either a beta function or a combination of Dirac delta functions is com-monly invoked to set a PDF of either combustion progress variable or mixture fraction, with the parameters of the PDF being determined based on its first and second moments computed by integrating proper balance equations. Because the choice of any of the above PDFs appears to be totally arbitrary as far as the underlying physics of turbulent combustion is concerned, the use of such PDFs implies weak sensitivity of the key averaged quantities to the PDF shape provided that different PDFs are characterized

by the same first and the same second moments. This implicit assumption was tested by compar-ing mean heat-release rates, burning velocities, etc. averaged by invoking either the beta-func-tion or double-Dirac-delta-function PDF, with all other things being equal [175].

Finally, the temperature and density of com-bustion products computed for gasoline-air mixtures by running the CHEMKIN software package under a wide range of conditions (vari-ous equivalence ratios, elevated pressures and unburned gas temperatures) were parameter-ized, and the parameterizations were imple-mented into OpenFOAM.

ResultsThe results of the turbulent combustion model development are briefly summarized above and are discussed in detail in the cited papers [173-175]. Here, we restrict ourselves to numerical results obtained by running OpenFOAM.

Results of numerous simulations [165-177] per-formed by running OpenFOAM with various spray models indicate that (i) a combination of the Rosin-Rammler distribution with Reitz-Diwakar secondary breakup model [247,248] and (ii) the properly implemented KHRT model [249-253] yield the best agreement with the measured [96] liquid penetration length and Sauter mean diameter (SMD) under the condi-tions of the present study, with the latter model showing the best performance as far as the SMD obtained from high-pressure (200 bar) sprays is concerned. Certain results computed using the KHRT model are plotted in lines in Figures 45 and 46, while symbols show experimental data by Hemdal et al. [96].

When running the original version of the OpenFOAM to simulate premixed turbulent combustion, it was observed that the mean temperature and density computed by the code did not agree with the widely recognized BML equations [254,255]. The source of the problem was found and the BML method for computing the mean temperature and density was imple-mented into the code.

Figure 46. Comparison of measured (symbols) and calculated (lines) liq-uid penetration for different ambient Ta and fuel Tf temperatures. Black circles and solid lines: Ta=350 K, Tf=243 K; red squares and dashed lines: Ta=295 K, Tf=295 K; blue plus-es and dot-dashed lines: Ta=350 K, Tf=320 K; magenta stars and dotted lines: Ta =295 K, Tf =243 K.

Figure 47. Comparison of calculated combustion temperatures and densi-ties computed by running EQUIL code (solid lines) and calculated using the approximations given by Eqs. (1)-(4) for different equivalence ratios at p=28 atm. Equivalence ratios are increased along the arrows.

Moreover, it was found that OpenFOAM over-estimates the temperature of stoichiometric combustion products, because it is computed by assuming the complete conversion of gasoline to CO2 and H2O. To resolve the problem, the temperature Tb and density ρb of equilibrium combustion products were computed using the EQUIL code in the CHEMKIN-II package for various equivalence ratios, unburned tempera-tures, and pressures. The results were approxi-mated as follows:

(1)

(2)

(3)

where subscript 0 refers to standard conditions (Tu,0=300 K, p0=1 atm). Dependencies of the coefficients αtb, βtb, γtb, αρu, αρb and βρb on the equivalence ratio φ have been approximated as follows:

(4)

Figure 45. Comparison of measured (symbols) and calculated (lines) liq-uid penetration and SMD for different injection pressures.Black circles and solid lines: pinj=50 bar, gasoline; red squares and dif-ferent injection pressures pinj. Black circles and solid lines: pinj=50 bar, gasoline; red squares and dashed line in subfigure (a): pinj=125 bar, gasoline; red squares and dashed line in subfigure (b): pinj=200 bar, ethanol; blue pluses and dot-dashed lines: pinj=200 bar, gasoline.

a) Liquid penetration b) SMD

Good agreement between Tb and ρb calculated by running EQUIL, represented by Eqs. (1)-(4), was obtained (see Figure 47) and these approxi-mations were implemented into OpenFOAM. This extension of the code has resulted in more realistic computed temperatures (see Figure 48a), and it has resolved the problem of non-physical, negative mass fraction of unburned gas that was produced by the original code (see Figure 48b). These improvements have signifi-cantly improved convergence of the numerical solutions.

Shown in Figure 49 are the spatial fields of the Favre-averaged regress variable, computed at different crank angle degrees by running OpenFOAM with the basic version of the Chalmers model of premixed turbulent com-bustion implemented into it.

Finally, the licentiate thesis was defended by Chen Huang [165].

a) Gasoline b) Ethanol

a) burned temperature for 0.2≤φ≤1.0

b) burned density for 0.2≤φ≤1.0

c) burned temperature for 1.2≤φ≤2.0

d) burned density for 1.2≤φ≤2.0


Conclusions and Ongoing WorkBy comparing results of experimental [96] and numerical studies of hollow-cone gasoline and ethanol sprays discharged by a piezo-actuated outward-opening pintle injector in Chalmers spray chamber, the properly implemented KHRT model was ranked best under the con-ditions of the measurements.

The widely recognized BML method [254,255] for evaluating the mean density and temperature in a premixed turbulent flame was implemented into OpenFOAM and the extended code was debugged. The basic version of the Chalmers model of premixed turbulent combustion was implemented into OpenFOAM and the extended code was debugged.

Approximations of the temperature and density of equilibrium products of combustion of gas-oline-air mixtures under various equivalence

Figure 48. Comparison of the Favre-averaged temperatures a) and mass fractions of the unburned mixture b) obtained for the standard (solid line) and implemented (dashed lines) li-braries for calculating the mean tem-perature and mean density within a premixed turbulent flame.

Figure 49. Spatial fields (at the plane where spark plug is located) of the Favre-averaged regress variable b, computed at different crank angle degrees using the basic version of the Chalmers model of premixed tur-bulent combustion.

ratios, elevated temperatures and pressures were implemented into OpenFOAM.

A simple algebraic model [172] of turbulent sca-lar transport in premixed flames was extended to be compatible with the Chalmers model of premixed turbulent combustion and the extended model was validated [173].

The presumed beta function PDF approach was confirmed [175] to be a reasonable tool for mod-eling the influence of turbulent fluctuations in unburned mixture composition on stratified and non-premixed combustion. However, a presumed PDF does not seem to be a reason-able tool to simulate the influence of turbulent fluctuations in the combustion progress variable on premixed or partially premixed turbulent combustion [175].

a) temperature b) mass fraction of the unburned mixture

The next-year plans are• to submit a journal paper,• to complete implementation of the Chalmers

model of turbulent combustion into OpenFOAM and to debug the extended code,

• to approximate the computed data base on the laminar flame speeds of gasoline-air mixtures characterized by various equiva-lence ratios under elevated temperatures and pressures associated with combustion in SI engines,

• to combine the Chalmers model of premixed turbulent combustion with the mixture-fraction balance equation and the beta-func-tion presumed PDF approach to simulating non-premixed turbulent combustion; to implement the combined model into OpenFOAM; and to debug the extended code in order to simulate strongly stratified flames, including mixing-controlled afterburning.

The long-term plans are• to combine the above models with flamelet

library approach; to implement the combined model into OpenFOAM; and to debug the extended code in order to simulate emissions from gasoline direct injection (GDI) engines,

• to apply the extended code to unsteady 3D RANS simulations of mixture formation and combustion in a GDI engine,

• to compare numerical results with experi-mental data obtained within the framework of the CERC project entitled “Spray-Guided Gasoline Direct Injection”,

• to complete the Ph.D. thesis by Chen Huang.

Figure 50 (to the left). Effect of turbu-lence model on the Favre-averaged regress variable, integrated over the combustion chamber. Black solid line: standard k-ε model; red dashed line: RNG k-ε model; blue dot-dashed line: k-ω SST model.

Figure 51 (to the right). Effect of varying Courant number Co on the Favre-averaged regress variable, integrated over the combustion chamber.

Project leaderAssoc. Prof. Mats Andersson

Optical Methods for Spray and Combustion Diagnostics

of scattered light, fluorescence or other induced emission. Usually pulsed lasers with a high peak power are used, combined with intensified CCD cameras which enable signal amplification and short exposure time. Thus, imaging can be accomplished even with low-concentration species and in luminescent environments such as flames. By proper choice of the wavelength of the laser light and the detected light vari-ous species can be selectively probed. In addi-tion, elastically scattered light from fuel drops (Mie scattering) can be detected to visualize the distribution of liquid fuel drops in sprays. In order to image also the distribution of vapor-phase fuel, laser-induced fluorescence (LIF) can be used. In that case the fluorescence light from fuel molecules in both liquid and vapor phase is detected at a longer wavelength than the excitation light. For this purpose a fuel with well controlled properties containing a specific fluorescent tracer molecule is used for optimum accuracy or selectivity of the measurement, but the fluorescent properties of molecules in com-mercial fuels can also be used. The fact that the absorption and fluorescence properties of many molecules are temperature dependent enables temperature measurements using LIF.

Besides the laser-based techniques, important information on combustion processes can be obtained by measuring the natural lumines-cence from the f lame. Since various mol-ecules emit chemiluminescence at different wavelengths, the emission spectrum contains information on the chemical reactions going on which can be combined with spatially and/or

Objectives The objectives of this project are to develop and apply optical measurement techniques for spray and combustion diagnostics. This includes spray diagnostics with the purpose to measure the distribution and concentration of liquid and vapor phase fuel, air fuel mixing, and temperatures. The combustion characteriza-tion methods aim to identify ignition, visualize flame propagation, and measure combustion intermediates and products. The focus is on developing methods that can find use in various CERC projects and to carry out investigations working with other CERC projects.

Background Optical measurement techniques have many applications in combustion engine research, and are widely used in both university and indus-try labs [256,243,257]. Optical techniques have several advantages that make them well suited to measure a number of properties, which are difficult or impossible to investigate by other means. Optical diagnostics is non-intrusive, but requires that light can be inserted into and/or collected from engines or spray chambers. The use of advanced optics, cameras and lasers enable measurements with high temporal, spa-tial and spectral resolution.

One particularly powerful concept is planar laser-sheet imaging. The cross-section of an object (spray, flame, combustion chamber etc.) is illuminated with the light from a laser formed into a thin sheet by expansion in one dimen-sion and focusing in the other. Perpendicular to the laser sheet a camera captures the image


temporally resolved detection. Because of the relative simplicity of chemiluminescence detec-tion, it is a good complement to the laser-based techniques and may be used in applications and experimental configurations where it is difficult to apply laser illumination.

The techniques described above are all well established and used in many labs, but that does not mean that there is no need for fur-ther development. On the contrary, there is intense activity aimed at improving existing techniques, inventing new ones, combining techniques in innovative ways and making use of new instrument technology. Furthermore, different measurement objects and situations may require that existing practices and methods need to be modified or improved to be applicable or to extract as much information as possible. Based on these considerations, the role of this project is to enable and assist CERC researchers to apply advanced optical measurement tech-niques at an appropriate level of complexity, to make the best use of the instrumentation avail-able for advanced spray and combustion engine experiments.

Methods Several different techniques have been evalu-ated and applied in various measurement applications in spray chambers or in engines with optical access. Like previous years, fluo-rescence-based techniques have been used for several applications. The detection of chemilu-minescence has also been used this year.

Combinations of planar laser-induced fluores-cence and Mie scattering have been used to image the distribution of liquid fuel and fuel vapor in cross sections of sprays. For the inves-tigation of gasoline-like fuels a five-component model fuel has been used to obtain a distillation curve similar to that of commercial gasoline. To this non-fluorescing model fuel various fluorescent tracers, representing fuel fractions with different volatility, are typically added. Ketones of different volatility (acetone, 3-pen-tanone and methylcyclohexanone) are used to trace the low-, medium-, and high-boiling-point fractions, respectively. A cross-section of the spray is illuminated with laser light formed into a thin sheet, and by using two cameras to detect the Mie scattered light and fluorescence light, respectively, the spatial distribution of fuel drops and the traced fuel component can be analyzed. Furthermore, fuel mixtures with two different tracers with different volatility and different spectroscopic properties can be used to simultaneously visualize the distribu-tion of light and heavy fuel fractions by detect-ing fluorescence light at different wavelength ranges by the two cameras.

Laser-induced fluorescence has also been used to image the distribution of NO molecules in a combusting spray [177]. In this case the light from a tunable dye laser, at a wavelength of ~226 nm was used for the excitation and detec-tion by an intensified CCD-camera was done at 248±5 nm to reduce the background of scattered laser light and flame luminescence. Measurements were carried out in the high-pressure/high-temperature spray chamber in a spray of dimethyl ether (DME) injected into air at 50 bar and 835 K.

In collaboration with the SGDI project, previ-ous measurements in the optical single cylinder engine [154] have been extended using a differ-ent type of fuel injector and the luminescence from combustion species and soot has been investigated. The focus of the recent measure-ments was to make a spatially and spectrally resolved detection of the luminescence to identify the formation of various combustion species.

Results Sprays from an outward-opening piezo-actuated injector have been analyzed with a combination of planar LIF and Mie scattering in the high-pressure/high-temperature spray chamber at conditions similar to those for late direct injection at cold start in a spark-ignited engine, i.e. temperature in the range 363-473 K. Under these conditions, the hollow-cone spray produces a torus-shaped fuel cloud with two counter-rotating vortices. This work [176] is an extension of previously published work [101,108], now also including a statistical analy-sis and further combinations of fluorescence tracers. Simultaneously recorded planar Mie and LIF images were used to analyze the evapo-ration of fuel and various fuel components in the hollow-cone sprays. Figure 52 shows two pairs of Mie and LIF images recorded at the same conditions, but with two different fluo-rescent tracers, acetone and methylcyclohexa-none (mch), tracing the light and heavy gasoline components respectively. The two images in the pair recorded with mch as tracer are almost identical, showing that the fluorescent tracer has the same spatial distribution as the fuel drops indicating that no evaporation of the heavy fuel component. However, in the other image pair with acetone as tracer there are clear differences between the two images, showing that fuel has evaporated and moved to different parts of the spray compared to the fuel drops. In order to facilitate the comparison between Mie and LIF image pairs, normalized difference images are calculated. The procedure for this begins by scaling the total intensity in the two images to match them, then the Mie image is subtracted from the LIF image and the difference is divided by the sum of the two images. Thus, an image

showing the relative difference is obtained. Figure 53 shows difference images calculated from the image pairs presented in Figure 52, where green color indicates no or very small difference, yellow-red color an enhanced vapor density and blue color an enhanced drop pres-ence. For the mch tracer the differences are small (mainly green color) with minor varia-tions only in areas with low fuel concentration, whereas areas with high tracer (yellow-red) and drop (blue) density are easily identified for the acetone-traced spray. As already seen in Figure 52, but further emphasized in Figure 53, the fuel vapor has primarily moved into the center of the vortices. Areas with an enhanced presence of drops are found at the top and the bottom of the fuel cloud. Furthermore, it can be noted that the vapor does not expand outside of the torus-shaped cloud generated by break-up of the hollow-cone spray, the vapor rather penetrates inwards to the centre of the expanding vortices.

In order to obtain an overview of the behavior of different fuel components, at different tem-peratures and at different times after start of injection a statistical analysis has been carried out. The number of pixels with a difference in certain intervals can be plotted in a histogram, as shown in Figure 54. In images with small dif-ferences, most pixels are in the central columns, whereas the distribution gets broader with more evaporation. Furthermore, the average of the absolute values of the differences in all pixels in the spray can be calculated with numbers

Figure 52. Pairs of simultaneously recorded Mie (left) and LIF (right) im-ages captured 1.5 ms asoi at an air temperature of 363K. The upper pair is recorded with methylcyclohexa-none as fluorescence tracer, and the lower pair with acetone. Image area 48x37 mm.

Figure 53. Normalized difference images for the two image pairs presented in Figure 52. The injector tip, and areas outside the spray are excluded from the calculation.

indicated in the upper right corners of the dia-grams in Figure 54. Thus, the histograms and the average differences can be used as a qualita-tive measure of the degree of evaporation and spatial separation of the vapor and the remain-ing drops. Average differences as a function of time after start of injection (asoi) are shown in


Figure 55 for acetone- and mch-traced sprays for three different air temperatures. Higher temperature clearly means faster evaporation, as can be expected. For the light fuel fraction there is a relatively rapid evaporation at all three temperatures, whereas the evaporation of the heavy components is almost non-existent at 363 K, and then the evaporation rate successively increases at 413 and 473 K, albeit still signifi-cantly slower than for the light fuel fraction.

The evaporation and resulting spatial distribu-tion of fuel fractions of different volatility was further investigated by adding two fluorescent tracer molecules to the model fuel, in this case acetone to trace the light fraction and cumene (isopropylbenzene) to trace the heavy fuel frac-tion. Both tracer molecules can be excited at the same wavelength, 266 nm, but emit fluores-cence at different wavelengths, cumene around

Figure 56. Image pairs of cumene fluorescence (left column) and ac-etone fluorescence (middle column) and normalized difference images with red color indicating a high ac-etone presence and blue color a high cumene fluorescence (right column). The images are recorded 2.5 ms asoi at the temperature of 363 K (top row), 413 K (middle row) and 473 K (bottom row).

Figure 54. Histograms of the num-ber of pixels with normalized differ-ences in different 0.1 wide intervals between the minimum and maximum values of ±0.95 at 0.6, 1.0, 1.6 and 2.3 ms asoi for acetone as a tracer at 363 K. The average difference is indicated in the upper right corner.

Figure 55. Average difference as a function of time asoi with acetone (top) and methylcyclohexanone (bot-tom) as tracers. The curves represent three temperatures 363 K (blue), 413 K (green) and 473 K (red).

280 nm and acetone around 420 nm, with little spectral overlap. Thus, using a dichroic mirror and appropriate filters images of the distribution of the different tracers, and thereby the different fuel fractions, can be obtained in the same way as the simultaneously recorded Mie and LIF images. Examples of LIF images of the two tracers, together with difference images cal-culated according to a similar procedure as for the Mie/LIF image pairs, are shown in Figure 56. It is clear that in all cases there is a dif-ference in the distribution of the different fuel components, although the differences become smaller at higher temperatures with more rapid evaporation.

The distribution of NO in a combusting DME-spray was measured using LIF [177]. Due to limited pulse energy from the dye laser, the beam was not expanded into a sheet. Instead

the ~5 mm diameter beam was allowed to pass at various vertical distances (10-90 mm) from the injector nozzle, and images were recorded at different time after the start of injection (3-14 ms). During the most intense combustion, 4-8 ms asoi, there is a strong extinction of the laser light with only a small fraction being transmit-ted through the spray, likely due to absorption by hot combustion gases and/or particles. Figure 57 shows averaged images recorded at different heights and different time steps, and several observations can be made. There is a widen-ing of the spray with increasing distance from the nozzle, but the asymmetry in the images, in particular at long distances from the nozzle and intermediate time steps, is an effect of the extinction of the laser light, possibly combined with stronger absorption of the fluorescence light generated in the center of the spray com-pared to the edges. Besides this asymmetry, the NO is fairly evenly distributed along the spray cross section, at least in the averaged images. Also at different heights, the intensity (at the most intense regions) is similar. Finally, it is noted that the NO fluorescence remains at a high but decreasing level after the main combustion is finished, ~10 ms asoi.

Conclusions and ongoing work The combination of planer LIF and Mie scat-tering was successfully used to investigate the

evaporation of different fuel fractions in sprays of a gasoline-like model fuel. By identifying dif-ferences between the simultaneously recorded LIF and Mie images, two types of conclusions could be made. First, by analyzing the differ-ence between the images a qualitative measure of the evaporation of the traced fuel component can be made, and a comparison of the relative evaporation rate of different fuel components and at different temperatures can be made. Second, the comparison of individual image pairs enables the identification of the parts of the spray with an enhanced drop and vapor concentration, respectively. In the torus-shaped fuel cloud with two internal counter-rotating vortices it was found that the vapor tends to accumulate in the centre of the vortices while there is a high drop concentration at the top and the bottom of the fuel cloud. Besides using Mie/LIF image pairs to compare the spatial distri-bution of drops and various fuel components, direct comparison of the spatial distribution of fuel components with different volatility could be obtained by adding two fluorescent tracers of different volatility and different spectral char-acteristics, which enables simultaneous imag-ing of the two traced fuel fractions.

The investigation of NO by LIF was the first step to build up a competence to detect com-bustion species with LIF in the spray chamber,


Ingemar DenbrattProfessorApplied Mechanics/Div. of Comb.

Bo EgardtProfessorSignals and Systems

Tomas McKelveyProfessorSignals and Systems

Mark Linne*Prof./Director CERC/Applied Mechanics/Div. of Comb.

Valeri GolovitchevAssoc. Prof.Applied Mechanics/Div. of Comb.

Sven AnderssonAssoc. Prof.Applied Mechanics/Div. of Comb.

Andrei LipatnikovAssoc. Prof.Applied Mechanics/Div. of Comb.

Senior StaffPetter DahlanderAssoc. Prof.Applied Mechanics/Div. of Comb.

Mats AnderssonAssoc. Prof.Applied Mechanics/Div. of Comb.

Arjan Helmantel (20%)Asst. Prof.Applied Mechanics/Div. of Comb.

Raúl Ochoterena**Asst. Prof.Applied Mechanics/Div. of Comb.

Stina HemdalAsst. Prof.Applied Mechanics/Div. of Comb.

Ingemar AnderssonAsst. Prof.Signals and Systems

Anders KarlssonAdj. Prof.Volvo Technology

* Involved in management.

and was limited to the format of a M.Sc. thesis project. Still several valuable conclusions for the future work could be made. NO could success-fully be detected by LIF, but the signal intensity recorded by the camera was fairly low. In the flame of the combusting spray the attenuation of the laser light is a limiting factor. Furthermore, in the high-pressure environment the effects of quenching and line-broadening need to be taken into account.

In the future there will be continued work on development and application of LIF and related techniques for the investigation of fuel distribu-tion, fuel evaporation and temperature in spray chamber and engines with optical access. LIF will also be applied for imaging of the distribu-tion of combustion radicals. Future activities in the project also include further developing equipment and methods for detection and analy-sis of chemiluminescence.

Figure 57. Fluorescence images, averages of ten sprays, recorded at 10-90 mm below the nozzle and at 3-12 ms after start of injection [177]. The width of each image is 52 mm.

** Left in June.

During 2011 eight Ph.D. students and about twelve senior researchers from Applied Mechanics, Signals and Systems and Volvo Technology were engaged in the various CERC research projects.

Personnel researching and working at CERC in 2011 include:

Human resources


Revenues Total Cash In-kind

STEM 7 000 7 000 0

SAAB Powertrain AB 330 0 330 *

Scania CV AB 671 600 71 *

Volvo Powertrain AB 2 219 600 1 619 **

Volvo Car Corporation AB 2 100 600 1 500 **

Statoil A.S. 127 120 7 ***

ABB AB 1 228 890 338 ***

AB Volvo Penta 100 100

Hoerbiger Control Systems AB 50 50

Lantmännen Aspen Petroleum AB 270 250 20

Honda Research 0 0 0

Reaction Design 742 0 742 ****

Chalmers Univ. of Technology 7 100 1 500 5 600

Transfer from previous year 2 010 2 010

TOTAL 23 947 13 720 10 227

BUDGET 21 500 11 960 9 540

Table 2. Actual contributions from members 2011 (KSEK).

Finances during the period 2010-2013For the period 2010-2013 the budget following the agreement between the three parties STEM/Industry/Chalmers, given in Table 1, was established.

In the summary of the budget some of the revenues from the participating companies are “efforts in kind”.

Table 2 shows actual input of cash respectively “efforts in kind” for the participating companies during the year 2011.

In Table 3, the cost of activities at Chalmers during 2011 are given, distributed by cost categories.

Table 4 shows a summary of the project expenses for the year 2011.

Table 4. Summary of project expenses 2011 (KSEK).

Salaries 6 876

Lab costs 381

Equipment and supplies 527

Travels 359

Miscellaneous; IT, premises, overhead 4 864

TOTAL 13 007

Contribution from members 13 720

Transfer to next year 713

Table 3. Expenses at Chalmers 2011 (KSEK).

Chalmers In-kind

Project Salaries Lab cost Equipm. Travels Misc. Total Cash Budget Chalmers Industry Total

Modeling of spray formation, ignition and combustion in internal combustion engines 665 199 368 1 232 1 232 900 65 357 1 654

Spray-guided gasoline direct injection 1 018 100 29 27 715 1 889 1 889 2 479 962 630 3 481

Injection strategies for diesel engines 365 130 123 515 1 133 1 133 680 561 1 619 3 313

LDD engine combustion 250 7 156 413 413 800 194 200 807

Advanced laser-based methods 485 50 12 34 293 874 874 800 280 1 154

Combustion models for Bio fuel 709 25 361 1 095 1 095 1 020 135 385 1 615

Modeling DISI engines 743 12 61 381 1 197 1 197 1 120 150 1 347

Spray turbulence interaction 537 26 68 265 896 896 1 090 135 1 031

Aspen Fuel 150 74 224 224 250 246 20 490

Nanoparticles II 401 33 100 21 246 801 801 1 333 91 71 963

Altenative fuels 57 19 17 22 115 115 115 773 7 895

Diesel Engine Optimization 35 13 229 277 277 400 350 1 000 1 627

LDD engine control 666 68 34 313 1 081 1 081 1 000 258 338 1 677

Administration 795 59 926 1 780 1 780 1 300 300 2 080

Common project costs 1 100

TOTAL 6 876 381 527 359 4 864 13 007 13 007 13 287 5 600 4 627 22 134

Ph.D studentsMalin Ehleskog Ph.D. stud Volvo Power Train ABMarkus Grahn Ph.D. stud Volvo Car CorporationChen Huang Ph.D. stud Applied Mechanics/Div. of CombustionAnders Johansson Ph.D. stud Applied Mechanics/Div. of CombustionMonica Johansson Ph.D. stud Applied Mechanics/Div. of CombustionAnne Kösters Ph.D. stud Applied Mechanics/Div. of CombustionMikael Thor Ph.D. stud Signal and SystemsJunfeng Yang Ph.D. stud Applied Mechanics/Div. of Combustion

Research Engineers and TechniciansSavo Girja Ph.D. Applied Mechanics/Div. of CombustionAlf Magnusson Ph.D. Applied Mechanics/Div. of CombustionLars Jernquist Lic.Eng. Applied Mechanics/Div. of CombustionTorbjörn Sima M.Sc. Applied Mechanics/Div. of CombustionDaniel Härensten Eng. Applied Mechanics/Div. of CombustionAnders Mattsson Eng. Applied Mechanics/Div. of CombustionAllan Sognell Eng. Applied Mechanics/Div. of CombustionMorgan Svensson Techn. Applied Mechanics/Div. of CombustionJan Möller Techn. Applied MechanicsGöran Stigler Techn. Applied Mechanics

Management of CERCCERC is an independent unit with its own budget and ac-counting, within the Department of Applied Mechanics at Chalmers University of Technology. CERC’s activities are governed by a board of directors appointed by the President of Chalmers in consultation with the member companies. As the head of CERC, the director Mark Linne has had the overall responsibility for coordination within the center. The board consists of the chairman, one academic member and five representatives from the member companies.

Tommy BjörkqvistChairman of the Board for the Swedish Internal Combustion Engine Consortium (SICEC), including: CERC at Chalmers University of Technology, KCFP at the University of Lund, and CCGEx at the Royal Institute of Technology

Börje GrandinVolvo Car Corporation AB

Pär GustafssonABB Automation Products AB

Per Lange Scania CV AB

Lennart Skoogh SAAB Automobile Powertrain

Sören UddVolvo Powertrain AB

Anna DuBoisChalmers University of Technology

In addition, Bernt Gustafsson from the Swedish Energy Agency takes part in all discussions. Research at CERC is pursued as described in this annual report within refer-ence groups, and project results are presented directly to the CERC board.

Revenues 2010 2011 2012 2013 TOTAL

STEM 7 000 7 000 7 000 7 000 28 000

SAAB Powertrain AB* 1 200 1 200 1 200 1 200 4 800

Scania CV AB 1 200 1 200 1 200 1 200 4 800

Volvo Powertrain AB 1 200 1 200 1 200 1 200 4 800

Volvo Car Corporation AB 1 200 1 200 1 200 1 200 4 800

Statoil A.S. 180 180 180 180 720

ABB AB 1 410 1 410 1 410 1 410 5 640

AB Volvo Penta 160 160 160 160 640

Hoerbiger Control Systems AB 150 150 150 150 600

Lantmännen Aspen Petroleum AB 300 300 300 300 1 200

Honda Research** 400 400 400 400 1 600

Chalmers Univ. of Technology 7 100 7 100 7 100 7 100 28 400

TOTAL 21 500 21 500 21 500 21 500 86 000

Table 1. Total Incomes for 2010-2013 period (KSEK).

* SAAB Powertrain went into bankruptcy; no payments were made.

** The Honda Research facility was damaged in the Japanese earthquake and so they left CERC for now.

Comments on in-kind distributions:* Equipment for projects and consultations.** Industrial PhD student, equipment for projects and consultations.*** Consultations.**** Reaction Design donates several software licenses and host Junfeng Yang for 6 months in San Diego.


CERC Publications and Presentations 2006 - 20112006

1. Lipatnikov, A.N. and Chomiak, J., “Numerical Tests of a Measurement Method for Turbulent Burning Velocity in Stagnation Flames”, Combustion Science and Technology, 178, 1117-1151, 2006.

2. Nassos, S. , Elm, E., Svensson, E., Nilsson, M. , Boutonnet, M. and Järås, S.G., “Microemulsion prepared Ni catalysts supported on Cerium-Lanthanum Oxide for the Selective Catalytic Oxidation of ammonia in gasified biomass”, Applied Catalysis B: Environmental, 64, 96–102, 2006.

3. Magnusson A., Jedrzejowski S. and Andersson S., “Spray-wall interaction: Diesel fuels impinging on a tempered wall”, SAE paper 2006-01-1116, 2006.

4. Montorsi, L., Magnusson, A., Andersson, S., “A numerical and experimental study of diesel fuel sprays impinging on a temperature controlled wall”, SAE paper 2006-01-3333, 2006.

5. Frederiksson, J., Bergman, M., Golovitchev, V.I. and Denbratt, I., ”Modeling the Effect of Injection Schedule Change on Free Piston Engine Operation”, SAE paper 2006-01-0449, 2006.

6. Helmantel, A. and Denbratt, I., “HCCI Operation of a Passenger Car DI Diesel Engine with an Adjustable Valve Train”, SAE Paper no. 2006-01-0029, 2006.

7. Husberg, T., Manente, V., Ehleskog, R. and Andersson, S., ”Fuel Flow Impingement Measurements on Multi-Orifice Diesel Nozzles”, SAE Paper no. 2006-01-1552, 2006.

8. Ehleskog, R., Golovitchev, V. I., Denbratt, I., Andersson, S. and Rinaldini, C. A. “Experimental and Numerical Investigation of Split Injections at Low Load in an HDDI Diesel Engine Equipped with a Piezo Injector”, SAE Paper no. 2006-01-3433, 2006.

9. Alriksson, M. and Denbratt, I., “Low Temperature Combustion in a Heavy Duty Diesel Engine Using High Levels of EGR”, SAE Paper no. 2006-01-0075, 2006.

10. Fredriksson, J., Golovitchev, V.I. and Denbratt, I., ”Numerical Investigation of a Loop-Scavenged Two-Stroke Free Piston Engine”, JSAE Paper 20065455, JSAE Annual Spring Congress, 2006.

11. Sathiah, P. and Lipatnikov, A.N., “Numerical Modeling of Stationary but Developing Premixed Turbulent Flames”, ASME Paper GT2006-90916, 2006.

12. Magnusson, A., Jedrzejowski, S. and Andersson, S., “Comparison of Heat Transfer and Spray Characteristics for a Model Fuel and Standard Diesel Sprays Interacting with a Temperature-Controlled Wall”, THIESEL 2006, Sep 12-15, Valencia, ES, 2006.

13. Lipatnikov, A.N. and Chomiak, J., “Effects of Large-Scale Stretching on Turbulent Flame Speed”, Proceedings of the Fifth International Symposium on Turbulence, Heat and Mass Transfer, Hanjalic, K., Nagano, Y., and Jakirlic, S. ed.s, 2006.

14. Lipatnikov, A.N. and Chomiak, J., “On the Applicability of Coherent Flame Models to Simulations of Developing Turbulent Flames”, Proceedings of the Fifth International Symposium on Turbulence, Heat and Mass Transfer, Hanjalic, K., Nagano, Y., and Jakirlic, S. ed.s, 2006.

15. Lipatnikov, A.N., “Some Basic Problems of Modeling of Transient Premixed Turbulent Flames”, Pulsed and Continuous Detonations, Roy, G., Frolov, S., and Sinibaldi, J. ed.s, Moscow, Torus Press, pp. 33-38, 2006.

16. Larsson, M. and Denbratt, I., “Comparison of Conventional Diesel and Fischer-Tropsch Diesel Fuels for HCCI Combustion”, JSAE Paper no. 20065422, 2006.

17. Berntsson, A., Denbratt, I., “HCCI Combustion Using a Spark Ignited Stratified Charge”, JSAE Paper no. 20065424, 2006.

18. Cantore, G., De Marco, C. A., Montorsi, L., Paltrinieri, F. and Rinaldini, C. A., “Analysis of a HSDI Diesel Engine Intake System by Means of Multi-Dimensional Numerical Simulations: Influence of Non Uniform EGR Distribution”, 2006 Spring Technical Conference of the ASME Internal Combustion Engine Division proceedings, ICES2006 – 1359, 2006.

19. Golovitchev, V. I., Montorsi, L. and Denbratt, I., “Towards a New Type of the Hybrid Engine: Two-Stroke Free – Piston Compression Ignited Engine”, FISITA 2006 World Automotive Congress, F2006P421, 2006.

20. Kusaka, J., Montorsi, L., Golovitchev, V. I. and Denbratt, I., “A Numerical Simulation of Combustion In a Heavy Duty Diesel Engine”, FISITA 2006 World Automotive Congress, F2006P398, Yokohama, Japan, 2006.

21. Bergman, M., Fredriksson, J. and Golovitchev, V.I., “Performance and Emission Formation Evaluation of a Diesel-Fueled Free Piston Engine using CFD”, Paper no. 20065454, JSAE Annual Spring Congress, 2006.

22. Salsing, H., Rinaldini, C. A., Golovitchev, V. I. and Denbratt, I. “Performance and Emissions of a DME Diesel Engine Equipped with a Fuel Specific Common Rail System”, JSAE paper No: 20065425, 2006.

23. Golovitchev, V. I., Montorsi, L., Rinaldini, C. A. and Rosetti A., “CFD Combustion and Emission Formation Modeling for a HSDI Diesel Engine Using Detailed Chemistry”, 2006 Fall Technical Conference of the ASME Internal Combustion Engine Division proceedings, ICEF2006 – 1506, 2006.

24. Dahlander, P., Lindgren, R. and Denbratt, I., “High-Speed Photography and Phase Doppler Anemometry Measurements of Flash-Boiling Multi-Hole Injector Sprays for Spray-Guided Gasoline Direct Injection”, 10th International Conference on Liquid Atomization and Spray Systems, ICLASS, 2006.

25. Andersson, M., Hemdal, S., Persson, F. and Rosén, A, ”Temperature determination based on spectral shift of exciplex fluorescence”, Proc. ICLASS-2006, Paper ID ICLASS06-165, 2006.

26. Golovitchev, V. I., Montorsi, L., Rinaldini, C. A. and Rosetti, A. CFD Combustion and Emission Formation Modeling for a HSDI Diesel Engine Using Detailed Chemistry, ICEF paper No: ICEF2006-1506, 2006.

27. Kusaka, J., Montorsi, L., Golovitchev, V. I. and Denbratt, I. “Numerical Simulation of Combustion in a Heavy Duty Diesel Engine”, FISITA Paper no: F2006P398, 2006.

28. Golovitchev, V. I., Montorsi, L. and Denbratt, I. “Towards a New Type of Hybrid Engine: The Two-Stroke Free-Piston Compression Ignited Engine”, FISITA Paper no. F2006P421, 2006.

29. McKelvey, T. and Andersson, I., “System identification of the crankshaft dynamics in a 5 cylinder internal combustion engine”, Proc. 14th IFAC Symposium on System Identification, 2006.

30. Ochoterena R. L., Andersson M. and Andersson S., “Determination of soot size and concentration in optically dense sprays by optical methods”, 10th Combustion Generated Nanoparticles conference, ETH-Zürich, 2006.

31. McKelvey, T. and Andersson, I., “System identification of the crankshaft dynamics in a 5 cylinder internal combustion engine. In Proc. 14th IFAC Symposium on System Identification, 2006.

200732. Lipatnikov, A.N., “Scalar Transport in Self-Similar, Developing, Premixed, Turbulent Flames”, Combustion Science and Technology, 179, 91 – 115, 2007.

33. Sathiah, P. and Lipatnikov, A.N. “Effects of Turbulent Flame Speed Development and Axial Convective Waves on Oscillations of a Long Ducted Flame”, Combustion Science and Technology, 179, 1433 – 1449, 2007.

34. Golovitchev, V. I., Bergman, M. and Montorsi, L., “CFD Modeling of Diesel Oil and DMEPerformance in a Two-Stroke Free Piston Engine”, Combust. Sci. and Tech., 179, 417-436, 2007.

35. Nassos, S., Elm, E., Svensson, E., Boutonnet, M. and Järås, S.G., “The influence of Ni load and support material on catalysts for the selective catalytic oxidation of ammonia in gasified biomass”, Applied Catalysis B: Environmental, 74, Issues 1-2, 92-102, 2007.

36. Johansson, Å. , Hemdal, S., Andersson, M. and Rosén, A., “Determination of OH Number Densities Outside of a Platinum Catalyst Using Cavity Ringdown Spectroscopy”, Journal of Physical Chemistry A, 111, 6798-6805, 2007.

37. Golovitchev, V., Montorsi, L., and Denbratt, I., “Numerical Evaluation of a New Strategy of Emissions Reduction by Urea Direct Injection for Heavy Duty Diesel Engines”, Engineering Applications of Computational Fluid Mechanics, vol. 1, No. 3, pp.189-206, 2007.

38. Lipatnikov, A.N. and Chomiak, J., “Global Stretch Effects in Premixed Turbulent Combustion”, Proceedings of the Combustion Institute, 31, pp. 1361-1368, 2007.

39. Sathiah, P. and Lipatnikov, A.N., “Effects of Flame Development on Stationary Premixed Turbulent Combustion”, Proceedings of the Combustion Institute, 31, pp. 3115-3122, 2007.

40. Golovitchev, V., Montorsi, L., Calik, A. and Ergeneman, A.M., “Application of Dynamic –T Parametric Maps to 3D Detailed Chemistry Combustion Analysis in Diesel Engines”, Proceedings of VIIIth Congress on Engine Combustion Processes, Schriftenreihe Heft 7.1, pp. 371-382, 2007.

41. Larsson, M. and Denbratt, I., “An Experimental Investigation of Fischer-Tropsch Fuels in a Light-Duty Diesel Engine”, SAE Paper no. 2007-01-0030, 2007.

42. Larsson, M. and Denbratt, I., “Combustion of Fischer-Tropsch, RME and Conventional Fuels in a Heavy-Duty Diesel Engine”, SAE Paper no. 2007-01-4009, 2007.

43. Ehleskog, R., Ochoterena, R. L. and Andersson, S., “Effects of Multiple Injections on Engine Out Emission Levels including Particulate Mass from a HSDI Diesel Engine”, SAE Paper no. 2007-01-0910, 2007.

44. Montorsi, L., Golovitchev, V. I., Denbratt, I, Corcione, F. and Coppola, S. “Numerical Evaluation of Direct Injection of Urea as NOx Reduction Method for Heavy Duty Diesel Engines”, SAE Paper no. 2007-01-0909, 2007.

45. Golovitchev, V., Montorsi, L., Denbratt, I., Corcione, F. and Coppola, S. ”Numerical Evaluation of Direct Injection of Urea as NOx Reduction Method for Heavy Duty Diesel Engines”, SAE Paper no. 2007-01-0909, 2007.

46. Salsing, H. and Denbratt, I., ”Performance of a Heavy Duty DME Diesel Engine – an Experimental Study”, SAE Paper no. 2007-01-4167, 2007.

47. Golovitchev, V., Montorsi, L., Calik, A., and Milani, M., “The EGR Effects on Combustion Regimes in Compression Ignited Engines”, SAE Paper no. 2007-24-0040, 2007.

48. Bergman, M., and Golovitchev, V., “Application of Transient Temperature vs Equivalence Ratio Emission Maps to Engine Simulations”, SAE Paper no. 2007-01-1086, 2007.

49. Lif, A., Skoglundh, M., Gjirja, S. and Denbratt, I., ”Reduction of soot emissions when combusting water-in-diesel emulsion and microemulsion fuel in a direct injection diesel engine”, SAE Paper no. 2007-01-10762007, 2007.

50. Golovitchev, V., Montorsi, L., and Calik, A., “Analysis of Combustion Regimes in Compression Ignited Engines Using Parametric -T Dynamic Maps”, JSAE Paper no. 20077260, SAE 2007-01-1838, 2007.

51. Skogsberg, M., Dahlander, P. and Denbratt, I., “Spray shape and atomization quality of an outward-opening piezo gasoline DI injector”, SAE Paper no. 2007-01-1409, 2007.

52. Montorsi, L., Magnusson, A., Andersson, S. and Jedrzejowski, “Numerical and experimental analysis of the wall film thickness for diesel fuel sprays impinging on a temperature controlled wall”, SAE Paper no. 2008-01-0486, 2007.

53. Alriksson, M., Gjirja, S., Denbratt, I., “The Effect of Charge Air and Fuel Injection Parameters on Combustion with High Levels of EGR in a HDDI Single Cylinder Diesel Engine “ SAE Paper no. 2007-01-0914, 2007.

54. Lipatnikov, A.N.,”On Characterization of Turbulence in Premixed Flames”, Proceedings of the 21st International Colloquium on the Dynamics of Explosion and Reactive Systems, 2007.

55. Berntsson, A.W., Andersson, M., Dahl, D. and Denbratt, I., “LIF for OH in the Negative Valve Overlap of a HCCI Combustion Engine”, Spark Ignition Engine of the Future conference, 2007.

56. McKelvey, T., Andersson, I. and Thor, M., “Estimation of Combustion Information by Crankshaft Torque Sensing in an Internal Combustion Engine”, Proc. 2nd International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, 2007.

57. Kärrholm Peng, F., Nordin, P. A. N. and Weller, H., ”Modeling Injector Flow Including Cavitation Effects for Diesel Applications”, ASME Fluids Engineering Conference, 2007.

58. Skogsberg,M., Dahlander, P. and Denbratt, I., ”An experimental study of mixture preparation and combustion in an optical engine using a piezo-actuated injector”, Direkteinspritzung im Ottomotor IV, p.143-159, 2007.

200859. Larsson, S. and Andersson, I., “Self-optimising control of an SI-engine using a torque sensor,” Control Engineering Practice, vol. 16, May. 2008, pp. 505-514.

60. Lipatnikov, A.N., “Some Basic Issues of the Averaged G-Equation Approach to Premixed Turbulent Combustion Modeling”, The Open Thermodynamics Journal, 2, 53-58, 2008.

61. Lipatnikov, A.N.,”Conditionally averaged balance equations for modeling premixed turbulent combustion in flamelet regime”, Combustion and Flame, 152 (4), 529-547, 2008.

62. Lipatnikov, A.N., ”Can We Characterize Turbulence in Premixed Flames?”, 32nd International Symposium on Combustion, McGill University, Montreal, Canada, August 3-8, 2008. Abstracts of Symposium papers. The Combustion Institute, Pittsburgh(2008).

63. Ochoterena, R. and Andersson, S., “Time and spatially resolved temperature measurements of a combusting diesel spray impinging on a wall”, SAE International Journal of Fuel and Lubricants, 1, 970-983, 2008.

64. Dahlander, P. and Lindgren, R., “Multi-hole Injectors for DISI engines: Nozzle Hole Configuration Influence on Spray Formation”, Paper no. SAE 2008-01-0136, 2008.

65. Berntsson, A.W., Andersson, M., Dahl, D. and Denbratt, I., “A LIF-study of OH in the Negative Valve Overlap of a Spark-assisted HCCI Combustion Engine”, SAE Paper no. 2008-01-0037, 2008.

66. Ochoterena, R.L., Larsson, M., Andersson, S. and Denbratt, I., “Optical studies of spray development and combustion characterization of oxygenated and Fischer-Tropsch fuels”, SAE Paper no. 2008-01-1393, 2008.


67. Bergman, M., Fredriksson, J. and Golovitchev, V.I., “CFD-Based Optimization of a Diesel-fueled Free Piston Engine Prototype for Conventional and HCCI combustion,” SAE Paper 2008-01-2423, 2008.

68. Dahl, D., Denbratt, I. and Koopmans, L.”An Evaluation of Different Combustion Strategies for SI Engines in a Multi-Mode Combustion Engine”, SAE Paper no. 2008-01-0426 , 2008.

69. Kärrholm Peng, F., Tao, F. and Nordin, P. A. N.,”Three-Dimensional Simulation of Diesel Spray Ignition and Flame Lift-Off Using OpenFOAM and KIVA-3V CFD Codes”, SAE Paper no. 2008-01-0961, 2008.

70. Bergman, M. and Golovitchev, V.I., “Modification of a Diesel Oil Surrogate Model for 3D CFD Simulation of Conventional and HCCI Combustion” SAE Paper no. 2008-01-2410, 2008.

71. Andersson, I, McKelvey, T. and Thor, M., “Evaluation of a closed loop spark advance controller based on a torque sensor”, SAE Paper no. 2008-01-0987, 2008.

72. Helmantel, A., “Reduction of NOx Emissions from a Light Duty DI Diesel in Medium Load Conditions with High EGR Rates”, SAE Paper no. 2008-01-0643, 2008.

73. Ehleskog, R. and Ochoterena, R.L., “Soot evolution in multiple injection Diesel sprays, SAE Paper no. 2008-01-2470, 2008.

74. Salsing, H. and Denbratt, I., ”Performance of a Heavy Duty DME Engine – the Influence of Methanol and Water in the Fuel”, SAE Paper no. 2008-01-1391, 2008.

75. Golovitchev, V.I. and Rinaldini, C.A., “Development and Application of Gasoline/EtOH Combustion Mechanism: Modeling of Direct Injection Ethanol Boosted Gasoline Engine,” COMODIA Paper S12-C, 2008.

76. Dahlander, P., Gutkowski and Denbratt, I. “Visualization of fuel sprays for stratified cold starts in gasoline direct injection engines”, Proc. 22nd European Conference on Liquid Atomization and Spray Systems, 2008.

77. Kärrholm Peng, F. and Tao, F., ”On Performance of Advection Schemes in the Prediction of Diesel Spray and Fuel Vapour Distributions”, Proc. 22nd European Conference on Liquid Atomization and Spray Systems, 2008.

78. Andersson, M., Hemdal, S. and Ochoterena, R.L., “Temperature measurements using exciplex fluorescence with TMPD and methylnaphthalene as tracers”, Proc. 22nd European Conference on Liquid Atomization and Spray Systems, Paper ID ILASS-08-P-19, 2008.

79. Milani, M., Montorsi, L. and Golovitchev, V.I., “Combined Hydrogen Heat Steam and Power Generation System,” Proceedings of the 16th International Conference of the ISTVS, 2008.

80. Kärrholm Peng, F., “Numerical Modelling of Diesel Spray Injection, Turbulence Interaction and Combustion”, Ph.D thesis, Dept. of Applied Mechanics, Chalmers University of Technology, Göteborg, Sweden, 2008.

81. Husberg, T., Denbratt, I. and Karlsson, A., ”Analysis of Advanced Multiple Injection Strategies in a Heavy-Duty Diesel Engine using Optical Measurements and CFD-Simulations”, SAE Paper 2008-01-1328, 2008.

200982. Ochoterena R., “High speed shadowgraph and diffraction based imaging for spray characterisation and combustion studies”, SAE 2009-24-0034, 2009.

83. Lipatnikov, A.N.,”Can we characterize turbulence in premixed flames?”, Combustion and Flame, 156, 1242-1247, 2009.

84. Andersson I. and Eriksson, L.,”A parametric model for ionization current in a four stroke SI engine”, Journal of Dynamic Systems, Measurement and Control, 131(2):1-11, 2009.

85. Andersson, I., McKelvey, T. and Thor, M., “Evaluation of torque sensor based cylinder balancing in an SI engine”, SAE International Journal of Passenger Cars - Electronic and Electrical Systems, 365--371, 2009.

86. Sedarsky, D., Gord, J., Meyer, T. and Linne, M., “Fast-framing ballistic imaging of velocity in an aerated spray”, Optics Letters, Vol. 34, No. 18, 2748-2750, 2009.

87. Berrocal, E., Sedarsky, D. L., Paciaroni, M. E., Meglinski, I. V. and Linne, M. A., “Laser light scattering in turbid media Part II: Spatial and temporal analysis of individual scattering orders via Monte Carlo simulation”, Optics Express, Vol. 17, No. 16, 13792-13809, 2009.

88. Linne, M., Paciaroni, M., Berrocal, E. and Sedarsky, D., “Ballistic Imaging of Liquid Breakup Processes in Dense Sprays”, invited review article, Proceedings of the Combustion Institute, Vol. 32, pp. 2147-2161, 2009.

89. Nogenmyr, K., Bai, X.S., Fureby, C., Petersson, P. and Linne, M., “Large Eddy Simulation Study on the Dynamics of a Low Swirl Stratified Premixed Flame”, Combustion and Flame, Volume 156, Issue 1, pp 25-36, 2009.

90. Golovitchev, V. and Yang, J., “Construction of combustion models for rapeseed methyl ester bio-diesel fuel for internal combustion engine applications”, Biotechnology Advances, 27 (5), 641-655, 2009.

91. Helmantel, A., Golovitchev, V.I. and Yang, J., “Injection Strategy Optimization for a Light Duty DI Diesel Engine in Medium Load Conditions with High EGR Rates,” SAE Paper no. 2009-01-1441, 2009.

92. Thor, M., Andersson, I. and McKelvey, T., “Parameterized diesel engine heat release modeling for combustion phasing analysis”, SAE Paper no. 2009-01-0368, 2009.

93. Thor, M., Andersson, I. and McKelvey, T., “Modeling, identification, and separation of crankshaft dynamics in a light-duty diesel engine”, SAE Paper no. 2009-01-1798, 2009.

94. Dahl, D., Andersson, M., Berntsson, A., Denbratt, I. and Koopmans, L., ”Reducing Pressure Fluctuations at High Load by Means of Charge Stratification in HCCI Combustion with Negative Valve Overlap”, SAE Paper no. 2009-01-1785, 2009.

95. Salsing, H., Ochoterena, R. and Denbratt, I., “Performance of a Heavy Duty DME-Engine – The Influence of Nozzle Parameters on Combustion and Spray Development”, SAE Paper no. 2009-01-0841, 2009.

96. Hemdal, S., Wärnberg, J., Denbratt, I. and Dahlander, P., “Stratified Cold Start Sprays of Gasoline-Ethanol Blends”, SAE Paper no. 2009-01-1496, 2009.

97. Golovitchev, V., Yang, J. and Savo, G., “Modeling of Combustion and Emissions Formation in Heavy Duty Diesel Engine Fueled by RME and Diesel Oil”, SAE Paper no. 09ICE-0006, 2009.

98. Sedarsky, D. L., Gord, J., Meyer, T. and Linne, M.A., “Velocity imaging of fluid structures in the near field of an aerated spray”, ICLASS 2009, the 11th Triennial International Annual Conference on Liquid Atomization and Spray Systems, 2009.

99. Berrocal, E., Kristensson, E., Sedarsky, D. and Linne, M., “Analysis of the SLIPI technique for multiple scattering suppression in planar imaging of fuel sprays”, ICLASS 2009, the 11th Triennial International Annual Conference on Liquid Atomization and Spray Systems, 2009.

100. Sedarsky, D., Berrocal, E. and Linne, M., “Numerical analysis of ballistic imaging for revealing liquid breakup in dense sprays”, ICLASS 2009, the 11th Triennial International Annual Conference on Liquid Atomization and Spray Systems, 2009.

101. Andersson, M. and Wärnberg, J., “Application of laser-induced fluorescence for imaging sprays of model fuels emulating gasoline and gasoline/ethanol blends”, ICLASS 2009, the 11th Triennial International Annual Conference on Liquid Atomization and Spray Systems, Paper ID ICLASS2009-189, 2009.

102. Yang, J., Larsson, M. and Golovitchev, V., “Engine performance and emissions formation for RME and conventional diesel oil: a comparative study”, ASME Internal Combustion Engine Division Spring Technical Conference, pp. 1-11, 2009.

103. Wärnberg, J., Hemdal, S., Andersson, M., Dahlander, P. and Denbratt, I., “Ignitability of Hollow Cone Gasoline/Gasoline-Ethanol Fuel Sprays”, Proc. 18th Aachen Colloquium “Automobile and Engine Technology”, 2009.

104. Thor, M., Andersson, I. and McKelvey, T., “Estimation of Diesel Engine Combustion Phasing from Crankshaft Torque Data,” Proceedings of the 2009 IFAC Workshop on Engine and Powertrain Control, Simulation and Modeling, 2009.

105. Lipatnikov, A.N., ”Transient Behavior of Turbulent Scalar Transport in Premixed Flames”, THMT’09, Proceedings of the International Symposium Turbulence, Heat and Mass Transfer 6, pp. 1-12, 2009.

106. Lipatnikov, A.N., ”Simulations of Scalar Transport in Developing Turbulent Flames Solving a Conditioned Balance Equation”, Sixth Mediterranean Combustion Symposium, 2009.

107. Huang, C. , Golovitchev, V. and Lipatnikov, A.N., ”A Chemical Model of Gasoline/Ethanol Blends”, The First Joint Meeting of the Scandinavian-Nordic and French Sections of the Combustion Institute, 2009.

108. Wärnberg, J., Hemdal, S. and Andersson, M., ”Ignitability of hollow cone gasoline/gasoline-ethanol sprays”, Aachener Kolloquium, Fahrzeug- und Motorentechnik, pp. 413-450, 2009.

109. Ehleskog, M., Gjirja, S. and Denbratt, I., ”Effects of High Injection Pressure, EGR and Charge Air Puressure on Combustion and Emissions in an HD Single Cylinder Diesel Engine”, SAE Int. J. Engines 2(2): 341-354, 2009.

110. Hemdal, S., Wärnberg, J., Dahlander, P. and Denbratt, I., “Stratified Cold Start Sprays of Gasoline-Ethanol Blends”, SAE International Journal on Fuels and Lubrication, vol.2, issue 1, p. 683-696, 2009.

2010111. Andersson, M. and Rosen, A., “Adsorption and reactions of O-2 and D-2 on small free palladium clusters in a cluster-molecule scattering experiment”, Journal of Physics-Condensed Matter, 22 ( 33 ), 2010.

112. Golovitchev, V., Kärrholm Peng, F., “Ignition: New Application to combustion Studies”, Handbook of Combustion Vol.1: Fundamentals and Safety, 85-106, 2010.

113. Franzoni, F., Milani, M., Montorsi, L. and Golovitchev, V., “Combined hydrogen production and power generation from aluminum combustion with water: analysis of the concept”, International Jounal of Hydrogen Energy, 35 (4 ), 1548-1559, 2010.

114. Lipatnikov, A.N. and Chomiak, J., “Effects of premixed flames on turbulence and turbulent scalar transport”, Progress in Energy and Combustion Science, 36 (1), 1-102, 2010.

115. Lipatnikov, A.N., “Simulations of scalar transport in developing turbulent flames solving a conditioned balance equation”, Combustion Science and Technology, 182 , 405-421, 2010.

116. Sabel’nikov, V. and Lipatnikov, A.N., “Rigorous derivation of an unclosed mean G-equation for statistically 1D premixed turbulent flames”, International Journal of Spray and Combustion Dynamics, 2 ( 4 ), 301-324, 2010.

117. Ochoterena R., Lif, A., Nydén, M., Andersson, S. and Denbratt, I., “Optical studies of spray development and combustion of water-in-diesel emulsion and microemulsion fuels”, Fuel, 89, p.122-132, 2010.

118. Ehn, A., Høgh, J., Graczyk, M., Norrman, K., Montelius, L., Linne, M. and Mogensen, M., “Electrochemical Investigation of Model Solid Oxide Fuel Cells in H2/H2O and CO/CO2 atmospheres using Nickel Pattern Electrodes”, Journal of the Electrochemical Society, 157 No.11, B1588-B1596, 2010.

119. DeCaluwe, S. C., Bluhm, H., Zhang, C., El Gabaly, F., Grass, M., Liu, Z., Jackson, G. S., McDaniel, A. H., McCarty, K. F., Farrow, R. L., Linne, M. A., Hussain, Z. and Eichhorn, B. W., “In situ characterization of ceria oxidation states in high-temperature electrochemical cells with ambient pressure XPS”, Journal of Physical Chemistry C, 114, 19853-19861, 2010.

120. Sedarsky, D., Berrocal, E., Linne, M., “Numerical analysis of ballistic imaging for revealing liquid breakup in dense sprays”, Atomization and Sprays, vol. 20, no. 5 , 407-413, 2010.

121. Zhang, C., Grass, M., McDaniel, A. H., DeCaluwe, S. C., El Gabaly, F., Liu, Z., McCarty, K. F., Farrow, R. L., Linne, M. A., Hussain, Z., Jackson, G. S., Bluhm, H. and Eichhorn, B. W., “Measuring fundamental properties in operating solid oxide electrochemical cells by using in situ X-ray photoelectron spectroscopy”, Nature Materials, 9, 944-949, 2010.

122. Whaley, J. A., McDaniel, A. H., El Gabaly, F., Farrow, R. L., Grass, M. E., Hussain, Z., Liu, Z., Linne, M. A., Bluhm, H. and McCarty, K. F., “Fixture for in-situ and in-operando characterization of electrochemical devices in traditional vacuum systems”, Review of Scientific Instruments, 81, 086-104, 2010.

123. El Gabaly, F., McDaniel1, A. H., Grass, M., Liu, Z., Bluhm, H., Linne, M. A., Hussain, Z., Farrow, R. L. and McCarty, K. F., ”Measuring individual overpotentials in an operating solid-oxide electrochemical cell”, Physical Chemistry Chemical Physics, 12, 12138–12145, 2010.

124. Linne, M., Sedarsky, D., Meyer, T., Gord, J. and Carter, C. ,“Ballistic Imaging of the Flow in the Interior of the Near-Field of an Effervescent Spray”, special issue of Experiments in Fluids to honor the career of Jürgen Wolfrum on his 70th birthday, 49:4, 911–923, 2010.

125. Berrocal, E., Kristensson, E., Richter, M., Linne, M. and Aldén, M., ”Multiple Scattering Suppression in Planar Laser Imaging of Dense Sprays by Means of Structured Illumination”, Atomization and Sprays, 20(2), 133–139, 2010.

126. Sedarsky, D., Paciaroni, M., Linne, M., Petersson, P., Zelina, J., Gord, J. and Meyer, T., “Model Validation Data for the Breakup of a Liquid Jet in Crossflow”, Experiments in Fluids, 49:391–408, 2010.

127. Sabel’nikov, V. and Lipatnikov, A.N., “Averaging of flamelet-conditioned kinematic equation in turbulent reacting flows”, Turbulence: Theory, Types and Simulation, ( chapter 10 ), 2010.

128. Huang, C., Golovitchev, V., Lipatnikov, A.N., “Chemical Model of Gasoline-Ethanol Blends for Internal Combustion Engine Applications”, SAE Paper n. 2010-01-0543, 2010.

129. Imren, A., Golovitchev, V., Cem, S. and Gerado, V., “The Full Cycle HD diesel Engine Simulation using KIVA-4 Code”, SAE Powertrains, Fuels and Lubricants Meeting, Oct 25-27, 2010, San Diego, California, USA, 2010.

130. Linne, M. A., Sedarsky, D., Meyer, T., Gord, J., Carter, C., ”Ballistic Imaging in the Near-Field of an Effervescent Spray”, ILASS Europe 2010, 23rd Annual Conference on Liquid Atomization and Spray Systems, Brno, Czech Republic, September, 2010.

131. Sedarsky, D., Berrocal, E., Linne, M., ”Numerical investigation of discriminated forward scattering: optimizing imaging optics for dense sprays”, ILASS Europe 2010, 23rd Annual Conference on Liquid Atomization and Spray Systems, Brno, Czech Republic, September, 2010.

132. Lipatnikov, A.N., “Conditioned moments in premixed turbulent reacting flows”, presented at the 33rd International Symposium on Combustion, Beijing, China, August 1 – 6, 2010.


133. Lipatnikov, A.N., “A test of conditioned balance equation approach”, presented at the 33rd International Symposium on Combustion, Beijing, China, August 1 – 6, 2010.

134. Nogenmyr, K. J., Petersson, P., Bai, X.-S., Fureby, C., Collin, R., Lantz, A., Linne, M. and Aldén, M., ”Structure and stabilization mechanism of a stratified premixed low swirl flame”, presented at the 33rd International Symposium on Combustion, Beijing, China, August 1 – 6, 2010.

135. Kristensson, E., Berrocal, E., Wellander, R., Ritcher, M., Aldén and M., Linne, M., ”Structured illumination for 3D Mie imaging and 2D attenuation measurements in optically dense sprays”, presented at the 33rd International Symposium on Combustion, Beijing, China, August 1 – 6, 2010.

136. Huang, C., Golovitchev, V. and Lipatnikov, A.N., “A Semi-Detailed Chemical Mechanism of Combustion of Gasoline-Ethanol Blends”, 33nd International Symposium on Combustion, Tsinghua University, Beijing, China, Abstracts of Work-in-Progress Poster Presentations. The Combustion Institute, Pittsburgh, August 1-6, 2010.

137. Lipatnikov, A.N. and Sabel’nikov, V., “A Simple Model of Turbulent Scalar Flux in Premixed Flames”, 33nd International Symposium on Combustion, Tsinghua University, Beijing, China, Abstracts of Work-in-Progress Poster Presentations. The Combustion Institute, Pittsburgh, August 1-6, 2010.

138. Lipatnikov, A.N. and Sabel’nikov, V., “Does the Mean Speed of Points at a Random Flame Surface Characterize the Motion of the Mean Flame Surface? ”, 33nd International Symposium on Combustion, Tsinghua University, Beijing, China, Abstracts of Work-in-Progress Poster Presentations. The Combustion Institute, Pittsburgh, August 1-6, 2010.

139. Sabel’nikov, V. and Lipatnikov, A.N., “G-Equation and Scalar Transport in Turbulent Flames”, 33nd International Symposium on Combustion, Tsinghua University, Beijing, China, Abstracts of Work-in-Progress Poster Presentations. The Combustion Institute, Pittsburgh, August 1-6, 2010.

140. Sabel’nikov, V. and Lipatnikov, A.N., “A Simple Model of Turbulent Scalar Flux in Premixed Flames”, 33nd International Symposium on Combustion, Tsinghua University, Beijing, China, Abstracts of Work-in-Progress Poster Presentations. The Combustion Institute, Pittsburgh, August 1-6, 2010.

141. Magnusson, A., Begliatti, M., de Borja Hervás, F. and Andersson, M. “Characterization of Wall Film Formation from Impinging Diesel Fuel Sprays using LIF”, Ilass Europe, Brno, Czech Republic, 2010.

142. Ochoterena R. Li P., Vera-Hernández M. and Andersson S., “Influence of cavitation on atomisation at low pressures using up-scaled and transparent nozzles”, Ilass-Europe, Brno, Czech Republic, 2010.

143. Nilsson, M. and Egardt, B., “Comparing Recursive Estimators in the Presence of Unmodeled Dynamics”, IFAC Workshop on Adaptation and Learning in Control and Signal Processing, ALCOSP, 2010.

144. Larsson, V., Johannesson, L. and Egardt, B., “Impact of Trip Length Uncertainty on Optimal Discharging Strategies for PHEVs”, Proceedings of the 6th IFAC Symposium Advances in Automotive Control, 2010.

145. Lipatnikov, A.N., “Expansion of Spherical Flames in Turbulent Gas: Phenomenology and Modeling”, 2010 International Workshop on Turbulent Ignition and Flame Propagation”, National Central University, Taiwan, July 29-31, 2010.

146. Sabel’nikov, V., Lipatnikov, A.N., “A simple model for evaluating conditioned velocities in premixed turbulent flames”, Euromech Fluid Mechanics Conference – 8, Abstracts, Technical University of Munich, 2010.

147. Lipatnikov, A.N., “Some Basic Problems of Modeling Transient Premixed Turbulent Flames”, Deflagrative and Detonative Combustion, 2010.

148. “Some Viewpoints on Spray Measurements Using Synchrotron Radiation”, Workshop on Synchrotron Tools for Studies of Combustion and Energy Conversion (planning for the new MAX synchrotron in Lund), Lund University, Lund, Sweden, December 7 & 8, 2010.

149. “Ballistic imaging in atomizing sprays”, SAOT Workshop on Spray Diagnostics, to be presented to the Erlangen Graduate School in Advanced Optical Technologies at Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany, October 12 and 13, 2010.

150. “New diagnostics for multiphase flows”, plenary talk presented at the Workshop of the Swedish Industrial Association for Multiphase Flows, Stockholm, Sweden, June 16, 2010.

2011151. Kösters, A. and Karlsson, A., “A Comprehensive Numerical Study of Diesel Fuel Spray Formation with OpenFOAM,” SAE Paper 2011-01-0842, 2011.

152. Lipatnikov, A.N., “A test of conditioned balance equation approach”, Proceedings of the Combustion Institute, 33:1497-1504, 2011.

153. Lipatnikov, A.N., “Conditioned moments in premixed turbulent reacting flows”, Proceedings of the Combustion Institute, 33:1489-1496, 2011.

154. Hemdal, S., Andersson, M., Dahlander, P., Ochoterena, R. and Denbratt, I., “In-cylinder soot imaging and emissions of stratified combustion in a spark-ignited spray-guided direct-injection gasoline engine”, International Journal of Engine Research, vol. 12, issue 6, p.549-563, 2011.

155. Thor, M., Egardt, B., McKelvey, T., Andersson, I., “Estimation of Combustion Phasing Using the Combustion Net Torque Method,” Proceedings of the 18th IFAC World Congress, Milano, Italy, 11833-11838, 2011.

156. Thor, M., Egardt, B., McKelvey, T., Andersson, I., “Parameterized Diesel Engine Combustion Modeling for Torque Based Combustion Property Estimation,” Submitted to the SAE 2012 World Congress. Under review.

157. Andersson, I., Thor, M., McKelvey, T., “The torque ratio concept for combustion monitoring of internal combustion engines,” Submitted to Control Engineering Practice. Under review.

158. Grahn, M., Olsson, J-O, McKelvey, T., “A Diesel Engine Model for Dynamic Drive Cycle Simulations”, Presented at IFAC 18th World Congress, August 28-September 2, 2011. Proceedings of the 18th IFAC World Congress 18(1), 2011, doi: 10.3182/20110828-6-IT-1002.01160, 2011.

159. Grahn, M., Johansson, K., Vartia, C., McKelvey, T., “A Structure and Calibration Method for Data-driven Modeling of NOx and Soot Emissions from a Diesel Engine”, Manuscript under review for SAE World Congress 2012.

160. Johansson, M., Yang, J., Ochoterena, R., Gjirja, S. and Denbratt, I., “NOx and soot emissions trends for combustion of RME, SME, PME and Biodiesel blends of RME; engine and spray experiments combined with simulations”, to be published in Fuel journal, 2012.

161. Kösters, A., Golovitchev, V. and Karlsson, A., “A Numerical Study of the Effect of EGR on Flame Lift-off in N-Heptane Sprays Using a Novel PaSR Model Implemented in OpenFOAM,” SAE technical paper, 2012-01-0153, 2012.

162. “Numerical Analysis of NOx Formation Trends in Bio-diesel Combustion using Dynamic ϕ-T Parametric Maps”, presented at the JSAE/ SAE Fuels, Lubricants & Powertrains, Kyoto, Japan, Aug 29- Sep 2, 2011.

163. “3D CFD Modeling of a Biodiesel-Fueled Diesel Engine Based on a Detailed Chemical Mechanism”, submitted to SAE World Congress, Detroit, U.S.A, 2012.

164. Lipatnikov, A.N., “Fundamentals of premixed turbulent flames”, CRC Press, 2012.

165. Huang, C., ”Simulation of Combustion and Mixture Formation for Gasoline Direct Injection Engine Application”, Licentiate Thesis, 2011.

166. Huang, C., “Simulations of high-pressure hollow-cone sprays of gasoline and ethanol using OpenFOAM”, The Fifth European Combustion Meeting, Cardiff, June 29th – July ,1st, 2011.

167. Huang, C and Lipatnikov, A.N., “Modelling of gasoline and ethanol hollow-cone sprays using OpenFOAM”, SAE Paper 2011-01-1896, 2011.

168. Huang, C., Lipatnikov, A., Golovitchev, V., Hemdal, S., Wärnberg, J., Andersson, M., Dahlander, P., and Denbratt, I., “Gasoline Direct Injection - simulations and experiments”, ILASS2011, The 24th European Conference on Liquid Atomization and Spray Systems, Estoril, Portugal, September, 5-7, 2011.

169. Lipatnikov, A.N., “Conditioned moments in premixed turbulent reacting flows”, Proceedings of the Combustion Institute, 33:1489-1496, 2011.

170. Lipatnikov, A.N., “A test of conditioned balance equation approach”, Proceedings of the Combustion Institute, 33:1497-1504, 2011.

171. Lipatnikov, A.N., “Transient behavior of turbulent scalar transport in premixed flames”, Flow, Turbulence, and Combustion, 86:609-637, 2011.

172. Sabel’nikov, V.A. and Lipatnikov, A.N., “A simple model for evaluating conditioned velocities in premixed turbulent flames”, Combustion Science and Technology, 183:588-613, 2011.

173. Lipatnikov, A.N. and Sabel’nikov, V.A., “Transition from countergradient to gradient turbulent scalar transport in developing premixed turbulent flames”, Seventh Mediterranean Combustion Symposium, Chia Laguna, Cagliari, Sardinia, Italy, September 11-15, 2011. Proceedings. CD. 12 pp., 2011.

174. Sabel’nikov, V.A. and Lipatnikov, A.N., “Towards an extension of TFC model of premixed turbulent combustion”, Seventh Mediterranean Combustion Symposium, Chia Laguna, Cagliari, Sardinia, Italy, September 11-15, 2011. Proceedings. CD. 12 pp., 2011.

175. Lipatnikov, A.N. and Huang, C., “A simple test of presumed PDF approach to modeling partially premixed turbulent combustion in engines”, Submitted, 2012.

176. Andersson, M., Wärnberg, J., Hemdal, S., Dahlander, P. and Denbratt, I., “Evaporation of Gasoline-Like and Ethanol-Based Fuels in Hollow-Cone Sprays Investigated by Planar Laser-Induced Fluorescence and Mie Scattering”, JSAE Tech. Paper 20119201/SAE Tech. Paper 2011-01-1889, 2011.

177. Alvarez, C., “Nitric oxide measurements in DME flames using laser induced fluorescence”, M.Sc. thesis, Chalmers University of Technology, 2011.

178. Dahl, D. and Denbratt, I., “HCCI/SCCI load limits and stoichiometric operation in a multicylinder naturally aspirated spark ignition engine operated on gasoline and E85”, International Journal of Engine Research, 12 (1), 58-68, 2011.

179. Dahl, D., Andersson, M. and Denbratt, I.,“The Origin of Pressure Waves in High Load HCCI Combustion: A High-Speed Video Analysis”, Combustion Science and Technology, 183 (11), 1266-1281, 2011.

180. Linne, M., “Analysis of X-Ray Phase Contrast Imaging in Atomizing Sprays”, Experiments in Fluids, DOI 10.1007/s00348-011-1251-7, 2011.

181. Sedarsky, D, Berrocal, E. and Linne, M., “Quantitative image contrast enhancement in transillumination of scattering media”, Optics Express, 19 (3), 1866-1883 (2011). This paper was selected by the editors for further publication in the Virtual Journal for Biomedical Optics, 2011.

182. Kristensson, E., Araneo , L., Berrocal, E., Manin, J., Richter, M., Aldén, M. and Linne, M., “Structured Laser Illumination Planar Imaging, Part I: Analysis of multiple scattering suppression in scattering and fluorescing media”, Optics Express, 19 (14), 13647, 2011.

183. Nogenmyr, K-J., Petersson, P., Bai, X-S., Fureby, C., Collin, R., Lantz, A., Linne, M. and Aldén, M., ”Structure and stabilization mechanism of a stratified premixed low swirl flame”, Proceedings of the Combustion Institute, 33 (1), 1567-1574, 2011.

184. Kristensson, E., Berrocal, E., Wellander, R., Ritcher, M., Aldén, M. and Linne, M., ”Structured illumination for 3D Mie imaging and 2D attenuation measurements in optically dense sprays”, Proceedings of the Combustion Institute, 33 (1), 855-861, 2011.

185. Chakraborty, N. and Lipatnikov, A., “Statistics of conditional fluid velocity in the corrugated flamelets regime of turbulent premixed combustion: A Direct Numerical Simulation study”, Journal of Combustion, (Article ID 628208), 2011.

186. Lipatnikov, A., “Burning rate in impinging jet flames”, Journal of Combustion (737914), 11, 2011.

187. Lipatnikov, A., “Reply to comments by Zimont”, “Comment on: ‘Conditionally averaged balance equations for modeling premixed turbulent combustion in flamelet regime’ Reply”, Combustion and Flame, 158 (10), 2073-2074, 2011.

188. Magnusson, A. and Andersson, S., “An Experimental Study of Heat Transfer between Impinging Single Diesel Droplets and a Metal Wall using a Surface Mounted Height Adjustable Rapid Thermocouple”, ILASS2011; The 24th European Conference on Liquid Atomization and Spray Systems, Estoril, Portugal, September, 5-7, 2011.

189. Kösters, A., Karlsson, A., Ochoterena, R., Magnusson, A. and Andersson, S., “Diesel sprays – modeling and validation”, ILASS2011; The 24th European Conference on Liquid Atomization and Spray Systems, Estoril, Portugal, September, 5-7, 2011.

Invited Presentations 2011190. “Challenges of Measurements in Dense, Multiphase Environments”, invited plenary talk to the International Workshop on Frontiers in Synchrotron Tools for Studies of Combustion and Energy Conversion, Shanghai, China, October 15-18, 2011.

191. “X-Ray Phase Contrast Imaging in Atomizing Sprays”, selected as a ‘Hot Topic’ special presentation to the Gordon Conference on Laser Diagnostics in Combustion, Waterville Valley, NH, August 14-19, 2011.

192. “Studies on Atomizing Spray Breakup Mechanics”, M. Linne, International Seminar for the German SFB 686 consortium on Model Based Control of Homogenized Low Temperature Combustion, Bielefeld, Germany, July 21, 2011.

193. “Spray Research at Chalmers”, M. Linne, presented to INSA de Rouen and CORIA on the installation of Linne as a visiting professor at INSA, Rouen, France, April 14, 2011.

Other references(cited in the report)

194. Fansler,T.D., Drake, M. C., Stojkovic, B. and Rosalik, M. E., ”Local fuel concentration, ignition and combustion in a stratified charged spark ignited direct injection engine: spectroscopic, imaging and pressure based measurements”, International Journal on Engine Research, vol. 4, No. 2, 2003.

195. Dahms, R.N., Drake, M.C., Fansler, T.D., Kuo, T-W. and Peters, N., ”Understanding ignition processes in spray-guided gasoline engines using high-speed imaging and the extended spark-ignition model SparkCIMM.”,” Part A: Spark channel processes and the turbulent flame front propagation” and “Part B: Importance of molecular fuel properties in early flame propagation”, Combustion and Flame, 158, p. 2229-2260, 2011.

196. Sabathil, D., Koenigstein, A., Schaffener, P., Fritzsche, J. and Doehler, A., ”The Influence of DISI Engine Operating Parameters on Particule Number Emissions”, SAE paper 2011-01-0143, 2011.

197. Buri, S., Kubach, H. and Spicher, U., “Effects of increased injection pressures of up to 1000bar – opportunities in stratified operation in a direct-injection spark-ignition engine”, International Journal of Engine Research, vol.11, special issue, p.473-484, 2010.


Editor....................Mark Linne, CERC Layout....................................Invencity Print.........Tryckfolket AB, March 2012

198. Velji, A., Yeom, K., Wagner, U. and Spicher, U., “Investigation of the formation and oxidation of Soot Inside a Direct Injection Spark Ignition Engine Using Advanced Laser-Techniques”, SAE paper 2010-01-0352, 2010.

199. Stevens, E. and Steeper, R., “Piston Wetting in an Optical DISI Engine: Fuel Films, Pool Fires, and Soot Generation”, SAE paper 2001-01-1203, 2001.

200. Hedge, M., Weber, P., Gingrich, J., Alger, T. and Khalek, I., “Effect of EGR on Particle emissions from a GDI engine”, SAE paper 2011-01-0636, 2011.

201. Alger, T., Chauvet, T. and Dimitrova, Z., ”Synergies between High EGR Operation and GDI Systems”, SAE international Journal on Engines, vol. 1, issue 1 ,p. 101-114, 2008.

202. Atkinson, C., Allain, M. and Zhang, H., “Using Model-Based Rapid Transient Calibration to Reduce Fuel Consumption and Emissions in Diesel Engines,” SAE Technical Paper 2008-01-1365, 2008.

203. Sobel, J.R., Jeremiasson, J. and Wallin, C., “Instantaneous Crankshaft Torque Measurements in Cars,” SAE Technical Paper Series 960040, 1996.

204. Powell, J.D., “Engine Control Using Cylinder Pressure: Past, Present, and Future,” Journal of Dynamic Systems, Measurement, and Control, vol. 115, 1993, pp. 343-350.

205. Jeremiasson, J. and Wallin, C., “Balancing of Individual Cylinders in a V8 Diesel Engine Based on Crankshaft Torque Measurement,” SAE Technical Paper Series 981063, 1998.

206. Wallin, C., Gustavsson, L. and Donovan, M., “Engine Monitoring of a Formula 1 Racing Car Based on Direct Torque Measurement,” SAE Technical Paper Series 2002-01-0196, 2002.

207. Aulin, H., Tunestål, P., Johansson, T., Johansson, B., ” Extracting Cylinder Individual Combustion Data From a High Precision Torque Sensor,” Proceedings of ICEF2010, 2010 Fall Technical Conference of the ASME Internal Combustion Engine Division, 2010.

208. Vibe, I.I., “Semi-empirical expression for combustion rate in engines,” Proceedings of a Conference on Piston Engines, 2009.

209. De Ojeda, W., “Effect of Variable Valve Timing on Diesel Combustion Characteristics”, SAE Paper 2010-01-1124, 2010.

210. Murata, Y., Kusaka, J., Odaka, M., Daisho, Y., Kawano, D., Suzuki, H., Ishii, H. and Goto, Y., “Achievement of Medium Engine Speed and Load Premixed Diesel Combustion with Variable Valve Timing”, SAE paper 2006-01-0203, 2006.

211. Benajes, J., Serrano, J.R, Molina, S., and Novella, R., “Potential of Atkinson cycle combined with egr for pollutant control in a HD diesel engine”, Journal of Energy Conversion and Management, no 50, p 174-183, 2009.

212. Benajes, J., Molina, S., Martín, J., and Novella, R., “Effect of advancing the closing angle of the intake valves on diffusion-controlled combustion in a HD diesel engine”, Applied Thermal Engineering, no 29, p1947-1954, 2009.

213. Brahma, I., Sharp, M.C., Frazier, T.R., “Empirical Modeling of Transient Emissions and Transient Response for Transient Optimization”, SAE Technical Paper 2009-01-1508, 2009.

214. Arsie, I., Pianese, C., Sorrentino, M., “Control Parameters Optimization in Automotive Diesel Engines via Two Zone Modelling”, Advances in Automotive Control, Volume #5, Part #1, 2007.

215. Wahlström, J., ”Control of EGR and VGT for Emission Control and Pumping Work Minimization in Diesel Engines”, PhD Theses, Linköping University, 2009.

216. Kirchen, P., Boulouchos, K., “Development and Validation of a Phenomenological Mean Value Soot Model for Common-Rail Diesel Engines”, SAE Technical Paper 2009-01-1277, 2009.

217. Wilhelmsson, C., Tunestål, P., Johansson, B., Widd, A., and Johansson, R., “A Physical Two-Zone NOx Model Intended for Embedded Implementation”, SAE Technical Paper 2009-01-1509, 2009.

218. Wenzel, S.P., “Modellierung der Ruβ- und NOx-Emissionen des Dieselmotors”, Ph.D. thesis, Otto-von-Guericke-Universität, Magdeburg, 2006.

219. Bannister, C.D., “Quantifying the Effects of Biodiesel Blend Ratio, at Varying Ambient Temperatures, on Vehicle Performance and Emissions”, SAE paper no 2009-01-1893, 2009.

220. McCormick, R.L., “Effects of Biodiesel Blends on Vehicle Emissions”, National Renewable Energy Laboratory, NREL/MP-540-40554, 2006.

221. Mueller, C.J., “An Experimental Investigation of the Origin of Increased NOx Emissions When Fueling a Heavy-Duty Compression Engine with Soy Biodiesel”, SAE paper no 2009-01-1792, 2009.

222. Tat, M.E., “Measurements of Biodiesel Speed of Sound and its Impact on injection Timing”, National Renewable Energy Laboratory, NREL/MP-510-31462, 2003.

223. Zhang, Y., “Impact of Biodiesel on NOx Emissions in a Common Rail Direct Injection Diesel Engine”, Energy and Fuels, vol 21, pages 2003-2012, 2007.

224. Karlsson, A., “Modeling Auto-Ignition, Flame Propagation and Combustion in Non-Stationary Turbulent Sprays”, Ph.D. Thesis, Dept. of Thermo- and Fluid Dynamics, Chalmers University of Technology, Göteborg, Sweden, 1995.

225. Nordin, N., “Complex Chemistry Modeling of Dieselspray Combustion”, PhD thesis, Dept. of Thermo- and Fluid Dynamics, Chalmers University of Technolgy, Göteborg, Sweden, 2001.

226. www.openfoam.com.

227. Kokjohn, S.L., Reitz, R.D., “Investigation of the Roles of Flame Propagation, Turbulent Mixing, and Volumetric Heat Release in Conventional and Low Temperature Diesel Combustion”, ASME J. Eng. Gas Turbines Power, Vol 133, 2011.

228. Hong, S., Wooldridge, M.S., Assanis, D.N., “Modeling of Chemical and Mixing Effects on Methane Autoignition Under Direct-Injection, Stratified Charged Conditions”, Proceedings of the Combustion Institute, 29:711-718, 2002.

229. Magnussen, B.F., “On the Structure of Turbulence and a Generalized Eddy Dissipation Concept For Chemical Reaction in Turbulent Flow,” 19th AIAA Sc. Meeting, St. Louis, USA, 1981.

230. Engine Combustion Network, www.sandia.gov/ecn.

231. Bogin, G.E. Jr., DeFilippo, A., Chen, J.Y., Chin, G., Luecke, J., Ratcliff, M.A., Zigler, B.T. and Dean, A.M., “Numerical and Experimental Investigation of n-Heptane Autoignition in the Ignition Quality Tester (IQT)”, Energy & Fuels, 2011.

232. Westbrook, C.K. et al., Combust. Flame (2010), doi:10.1016/j.combustflame.2010.10.020, 2010.

233. Naik, C.V., Puduppakkam, K.V., Liang, L. and Meeks, E., “Ignition Quality Tester Guided Improvements to Reaction Mechanisms for n Alkanes: n-Heptane to n-Hexadecane”, SAE World Congress 2012, 12PFL-0462 (submitted), 2012.

234. Murphy, M.J., “Compendium of Experimental Cetane Number Data”, NREL/SR-540-36805, 2004.

235. Reaction Workbench 15101: Reaction Design, San Diego, 2010.

236. Liang, L., Naik, C., Puduppakkam, K., Wang, C., Modak, E., Meeks, E., Ge, H-W, Reitz, R., and Rutland, C., “Efficient Simulation of Diesel Engine Combustion Using Realistic Chemical Kinetics in CFD”, SAE Technical Paper 2010-01-0178, 2010.

237. Ciezki, H.K. and Adomeit, G., Combust. Flame 93:421–433, 1993.

238. Pfahl, U., Fieweger, K. and Adomeit, G., Proc. Combust. Inst. 26:781–789, 1996.

239. Bax, S., Hakka, M.H.,Glaude, P.A., Herbinet, O. and Battin-Leclerc, F., Combust Flame 2010, 157, 1220–1229, 2010.

240. CHEMKIN-Pro software, Reaction Design Inc., MFC-GS-50-1012-UG-1, August 2010.

241. FORTÉ software, Reaction Design Inc., FOR-UG-40102-1009-UG-1b, September 2010.

242. Standard Test Method for Determination of Ignition Delay and Derived Cetane Number (DCN) of Diesel Fuel Oils by Combustion in a Constant Volume Chamber, ASTM D6890-11, ASTM International, 2011.

243. Drake, M.C. and Haworth, D.D. ,”Advanced gasoline engine development using optical diagnostics and numerical modeling”, Proceedings of the Combustion Institute, 31:99-124,2007.

244. Schwarz, C., Schunemann, E., Durst, B., Fischer, J. and Witt, A., “Potential of the spray-guided BMW DI combustion system”, SAE Paper 2006-01-1265, 2006.

245. Veynante, D. and Vervisch, L., “Turbulent combustion modeling”, Progress in Energy and Combustion Science. 28:193-266, 2002.

246. Lipatnikov, A.N. and Chomiak, J., “Turbulent flame speed and thickness: phenomenology, evaluation, and application in multi-dimensional simulations”, Progress in Energy and Combustion Science, 28:1-73, 2002.

247. Reitz, R.D. and Diwakar, R., “Effect of drop breakup on fuel sprays”, SAE Technical Paper 860469, 1986.

248. Reitz, R.D. and Diwakar, R., “Structure of high-pressure fuel sprays”, SAE Technical Paper 870598, 1987.

249. Reitz, R.D. and Bracco, F.V., “Mechanisms of breakup of round liquid jets”, In: Encyclopedia of Fluid Mechanics, Gulf Pub., NJ, vol. 3, pp.233-249, 1986.

250. Reitz, R.D., “Modeling atomization processes in high-pressure vaporizing sprays”, Atomization and Spray Technology, 3:309-337, 1987.

251. Su, T.F., Patterson, M.A., Reitz, R.D., and Farrell, P.V., “Experimental and numerical studies of high pressure multiple injection sprays”, SAE Technical Paper 960861, 1996.

252. Patterson, M.A. and Reitz, R.D., “Modeling the effects of fuel spray characteristics on Diesel engine combustion and emission”, SAE Technical Paper 980131, 1998.

253. Beale, J.C. and Reitz, R.D., “Modeling spray atomization with the Kelvin-Helmholtz/Rayleigh-Taylor model”, Atomization and Sprays, 9:623-650, 1999.

254. Bray, K.N.C. and Moss, J.B., “A unified statistical model for the premixed turbulent flame”, Acta Astronautica, 4:291-319, 1977.

255. Libby, P.A. and Bray, K.N.C., “Variable density effects in premixed turbulent flames”, AIAA Journal 15:1186-1193, 1977.

256. Kohse-Höinghaus, K. and Jeffries, J.B. (Eds), “Applied Combustion Diagnostics”, Taylor & Francis, New York, 2002.

257. Aldén, M., Bood, J., Li, Zh. and Richter, M., Proc. Comb. Inst. 33, 69, 2011.

CERCThe Combustion Engine Research Center (CERC) is a Swedish public/private Competence Center funded by the Swedish Energy Agency, Chalmers University, and industry in roughly equal parts. The aim of the center is to organize and focus relevant basic research on Internal Combustion Engines.

CERC goals are to reduce both of fuel consumption and engine exhaust emissions, with a view towards direct fuel injection, new combustion modes and new engine architectures. The center includes experimental theoretical projects. Activities include spark and compression ignited engine concepts, basic fuel spray combustion research, advanced control of engines, and evaluation of alternative fuels. Areas of strength in modeling and advanced diagnostics are brought to bear on these problems to advance the center’s research.

CERCChalmers University of TechnologySE-412 96 GöteborgSweden

Telephone: +46 (0)31-772 18 20E-mail: [email protected]: http://www.chalmers.se/am/cerc-en/

Date post:	21-Mar-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

CERC Annual Report 2011 - Chalmers · The three centers have a common board within SICEC, with a...

Documents