advanced combustion and emission control (acec) … · advanced combustion and emission control...
TRANSCRIPT
1
1
Advanced Combustion and Emission Control (ACEC)
Activities
February 8, 2005
2
FreedomCAR Advanced Combustion and Emission Control (ACEC) Goals
• Reduce petroleum dependence by removing critical technical barriers to the mass commercialization of high-efficiency, emissions-compliant internal combustion engine (ICE) powertrains
• Primary directions− ICE efficiency improvements for cars and light-duty trucks through low-
temperature combustion and minimization of thermal and parasitic losses − Emission control development integrated with combustion strategies for
emissions compliance and minimization of efficiency penalty− Increased efficiency and power density for hydrogen ICE− Coordination with fuels R&D to enable clean, high-efficiency engines using
both hydrocarbon-based fuel (petroleum and non-petroleum) and H2
• 2010 technical goals– Best engine brake thermal efficiency 45%– Powertrain cost $30/kW – NOx & PM emissions Tier II Bin5
2
3
ACEC Goals and Status
2004 Status Goals by Fiscal Year PFI DI 2007 2010 2013 2015
Goals for HC Fuel
ICE peak brake thermal efficiency % 30 41 43 45 46
ICE Powertrain cost (1) $/kW 20 30 35 30 30
Projected vehicle emissions Tier 2 < Bin 10 Bin 10 Bin 5 Bin 5 Bin 5
Emission control fuel economy penalty (2) % <5 <4 <3
Emissions Durability k miles 120 120 120 120 120 120
Goals for H2 Fuel
H2 ICE peak brake thermal efficiency % 38 45 45
H2 ICE Powertrain cost (1) $/kW 45 30
Projected vehicle emissions Tier 2 < Bin 5 Bin 5 Bin 5
(1) High volume production of 500,000 per year
(2) Fuel economy penalty over combined Federal Test Procedure due to emission control relative to diesel vehicle with 2003 emissions
Goals reflect strategy to move from gasoline PFI to diesel/gasoline/H2 DI
4
US Diesel Emission StandardsFuel Sulfur Regulation: current - 500 ppm, future - 15 ppm starting in 2006
0 2.0 4.00.2
0.01
0.1
NOx g/bhp-hr
PM g
/bhp
-hr
Fed. 1998Fed 2004
NOx+HC
Fed 2007-10
Heavy Duty Diesel Truck Emission Standards (>8500 lbs)
CA 2002
CurrentDiesel
P.M. g/ mi
NOx , g/ mi
Bin9
Bin10
EURO 4 Cars
EURO 3 Cars
Bin8
0.08
0.06
0.04
0.02
0
0.01
0.07 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Bin5
Diesel Tailpipe Emissions Regulations (< 8500 lb)
CurrentDiesel
• By 2009, same standards for all US cars & light-duty trucks • Fleet average of Bin 5, maximum emissions Bin 8• In states with California emissions, only Bin 5 diesels
• US standards are more stringent than current EU standards• HD diesels have emission challenges similar to LD• Paths to Bin 5:
• With existing engine-out emissions, must achieve 90% PM/NOx reduction• By further reducing engine-out emissions, aftertreatment burden reduced
3
5
Research StrategyPNGV FutureTodayEmission Control
Low TempCombustion
Combustion OptimizationReduce Thermal/Mechanical Losses
H2 ICE
Combustion & Emissions HC Fuels
Improved Efficiency
Enablers
Combustion Understanding
Direct InjectionCombustion Optimization
Modeling, kinetics, fundamentals
2003LTC at Light Load
2006Durable Bin 5 System
2010Bin 5 Emissions45% PeakEfficiency
201045% Peak Efficiency
2007Expanded Load Range
2005Sulfur Trap Feasibility
2007Combustion Control Sensors
2004PM Measurement Systems
2015$30/kW
2010$30/kW
2003Bin 5 Performance for Fresh Catalysts
EmissionsFor LTC
6
Combustion• Focus on new processes with potential for improved efficiency
and reduced emissions– Low Temperature Combustion (LTC) strategies such as Homogeneous
Charge Compression Ignition (HCCI).• Barriers
– Poor understanding of LTC fundamentals– LTC operating range limited to low to moderate loads– Control of combustion and heat-release rates for LTC for steady and
transient conditions• Strategy
– LTC to reduce engine-out emissions which reduce requirements for aftertreatment system
– Expand LTC to cover more of the driving schedule – Improve engine efficiency – Study interaction of fuel properties on LTC to determine synergies– Improve understanding of combustion fundamentals and thermodynamic
processes– Synergy with hybrids
4
7
Combustion Research is Focused on Developing the Knowledge-Base Needed to Implement LTC.
Adiabatic flame temperaturein air
CO to CO2conversionstops
1000 1400 1800 2200 2600 30000
1
2
3
4
5
6
Temperature [K]
Equi
vale
nce
ratio
Soot li
mit
Soot
NOX
Diesel combustion– controlled heat release (mixing)– controlled combustion timing– wide load range– high efficiency (relative to Sl)– NOx and PM emissions
Akihama et. al 2001 SAE2001-01-0655
SI combustion (stoichiometric) – controlled heat release
(flame propagation) – controlled combustion timing– wide load range– three-way catalyst– throttling losses– lower efficiency
(relative to diesel)
HCCI
Low-Temperature Combustion– low NOx and particulate emissions– offers diesel-like efficiency – load range?– combustion timing?– heat release rate?– transient control?– robust combustion system?– fuel?
HCCI
8
1000 1400 1800 2200 2600 30000
1
2
3
4
5
6
Temperature [K]
Equi
vale
nce
ratio
Soot li
mit
Soot
NOX
• Importance & Status– New methods for LTC with diesel-like, mixing-controlled heat release were
discovered - one approach uses no EGR.– Since combustion is mixing-controlled, unpredictable heat-release rates in
premixed systems are avoided. • Next steps
– Determine the boundaries & robustness of mixing-controlled LTC – Investigate the feasibility of using in an engine.
Mixing-Controlled LTC Demonstrated in a Combustion Vessel (SNL)
8% 10%
21% O2
15%
- Higher mixing before combustion prevents NOx & soot formation.
- No EGR (21% O2).
- Low-temperature rich combustion prevents soot and NOx formation
- High EGR with less mixing
- Mixing before combustion prevents soot formation.
- EGR used to lower NOx.
Conventional diesel combustion
Axial distance from injector [mm]6040200
180 µm
Image of Flame Location(injector on left of figure)
5
9
• Importance– Late-injection LTC often sacrifices fuel
economy for low emissions.– Previously, kinetics largely thought to
dominate the entire combustion process. • Status
– Variable swirl alters the in-cylinder air motion and can produce counter-rotating vortices
• At optimal conditions, fuel and air meet at interface with high mixing and combustion phasing is improved
– Best fuel economy and lowest soot found between swirl ratios, Rs, of 2.5 and 3.5.
• Next steps– Investigate early injection strategies and
the potential of flow-structure enhanced mixing to extend the speed/load range.
In-Cylinder Flow Structures Promotes Air-Fuel Mixing in Late-Injection Diesel LTC (SNL and UW).
Fuel
Air
Rs = 1.44Non-optimal
Rs = 2.59Optimal
Fuel
Air
In-Cylinder Air Motion During Combustion at Two Swirl Levels
BoreCenter
CylinderWall
10
Achieving and Transitioning to Efficient Low Temperature Combustion (ORNL)
• Accomplishment– Achieved LTC with reduction of
engine-out NOx (90%) and soot (50%) at 20% load.
– Equal engine efficiencies during LTC and normal combustion.
– The transition between modes was several seconds. This was faster than other studies but not sufficient for a production vehicle. The ability to achieve the LTC condition depended on the transition path.
• Importance– The combustion process is not
degraded and HC emissions are not excessive.
• Next Steps– Demonstrate load expansion with
low-pressure EGR and partially premixed charge preparation.
– Demonstrate effects of transition path for expanding LTC operation range.
0
10
20
30
40
Torq
ue (f
t.lb)
0
20
40
60
80
EGR
(%)
0
50
100
150
200
NOx
(ppm
)
0
0.2
0.4
0.6
0.8
0 50 100 150 200 250 300
Time (s)
LII P
M (V
)
Base Efficient LTC BaseBase Efficient LTC BaseBSFC 212 g/hp•hr
Soot
(LII)
EGR
(%)
NO
x (p
pm)
Torq
ue (f
t-lb)
BSFC 212 g/hp•hrBSFC 212 g/hp•hr
1500 rpm, 2.6 bar bmep
6
11
University Consortia Developing LTC Optimization and Control Methods (UW, PSU, UM, UCB, MIT, Stanford)• Importance
– Engine control strategies a key to utilizing LTC– Computer simulation used for engine design and
evaluation of proposed control strategies
• Status– Gasoline HCCI transient behavior and mode changing
demonstrated in engine experiments.– Ignition, combustion and heat transfer submodels
developed and incorporated into GT-power.– Diesel LTC under load transients controlled with
injection and intake valve timing control.– Calculated fuel-spray patterns with variable-spray
angle produce minimal diesel fuel on wall and potential for increased combustion efficiency
• Next steps– Demonstrate strategies on laboratory engines– Extend simulation to full transient vehicle environment
Optimized spray angle in each injection pulse gives improved fuel-air mixing, reduced wall fuel
Star
t of C
ombu
stio
n
Intake Valve Closure (IVC) Timing
Controlled SOC, 3% efficiency change
High load -VVT & inj controlHigh loadLow load
Diesel Spray Pattern
Calculated Load Change for Diesel LTC
( 0 = TDC Compression )
12
Combustion Modeling and Chemical Kinetics are Keys to Understanding the Detailed Combustion
ProcessesThe computational fluid dynamics codes and kinetic data developed by National Labs are relevant to and used by the industry as a whole
1200 rpm, 100kPa, T(BDC): Experimental=409K, Model=407- 408K
356.4
356.6
356.8
357.0
357.2
357.4
357.6
357.8
358.0
0.04 0.08 0.12 0.16 0.2 0.24 0.28 0.32Equivalence Ratio
Cra
nk A
ngle
, 10%
Cum
. HR
[°]
Dec and SjobergModel: Mixture-1Model: Mixture-2Model: Mixture-3
Ignition Performance Predicted with Multi-Component Kinetic for Gasoline (LLNL)
Temperature Distribution in a Multi-Component Evaporating Fuel Spray Calculated by KIVA-4 (LANL)
7
13
0 µs
15 µs
29 µs
44 µs
59 µs
74 µs
88 µs
103 µs
118 µs
133 µs
Spray evolution at 500 bar of injection pressure under 5 bar of N2Colors show mass per unit area
X-Ray Measurements Reveal Structure of Diesel Spray Core (ANL)
Importance• X-ray absorption used for the first quantitative
measurements of the fuel-mass near the spray nozzle as functions of time and position
• Mass and momentum data improves the accuracy of spray models, enhancing their ability to reproduce and ultimately predict the structure of the spray under different combustion regimes
Next Steps• Extend scope to measurements under actual
engine conditions• Collaborate with spray modelers to improve
the accuracy and predictive power of spray simulations using unique near-nozzle data
• Use improved modeling to guide fuel-system development for lower emissions and higher efficiency
Value• These measurements were made with a
unique facility and personnel at ANL
14
Co-Development of Fuels and Advanced Combustion Engines is Essential to Ensure Availability of Fuels Optimized for New Combustion Regimes
Work Currently Underway• Characterize sensitivity of
advanced combustion regimes to variations in fuel properties
HCCI Combustion Phasing Affected by Fuel Properties
356357358359360361362363364365
0.04 0.08 0.12 0.16 0.2 0.24 0.28 0.32Equivalence Ratio [φ]
Cra
nk A
ngle
, 50%
Cum
. HR
Fire 19/1, PRF80Fire 19/1, iso-OctaneFire 19/1, Gasoline
Fuel-chemistry effects isolated
Future Plans• Through parametric analysis and modeling, identify relevance of fuel properties at a
molecular level to advanced combustion regimes
• Identify bulk fuel properties (e.g., an analog to cetane) that characterize fuels optimal for advanced combustion regime operation
8
15
H2 ICE Research• Objective
– Develop high-efficiency, low-emission H2 SI engines which bridge toward the future H2 economy
• Barriers− Low power density for ultra-low emission operation− Unknown combustion stability, pre-ignition and knock
characteristics − Lack of understanding of direct-injection mixing processes− NOx emissions at higher power density
• Strategy− Achieve power density comparable to gasoline PFI with direct
injection, pressure boosting, intercooling− Achieve bin 5 or lower emissions with a combination of lean and
stoichiometric combustion− Develop the fundamental knowledge-base and simulation tools − Other technical areas develop infrastructure and on-board storage
16
Emission Control Enables High-Efficiency, Lean-Burn Engines
• NOx aftertreatment approaches−LNT: lean NOx trap, also called NOx adsorber−Urea SCR: selective catalytic reduction with urea (ammonia)−HC SCR: selective catalytic reduction with hydrocarbons−Plasma: electrical energy creates ions to react with exhaust gases
• Barriers−LNT: fuel economy penalty, degradation by sulfur and high temp−Urea-SCR: infrastructure, enforcement −PM filter: fuel economy penalty from back pressure and effective
regeneration strategies• Strategy
−LNT: improve conversion at lower temperatures, determine degradation mechanisms due to poisons and ways to reverse effect
−Urea: system control to prevent breakthrough & unregulated emissions−PM filter: new regeneration methods−Modeling: improved fundamental understanding and simulations
9
17
Typical Emission Control System
Urea added
SCR/LNTcatalyst
HC, CO control NOx control
OxCat
PMfilter
PM control
NOx sensor NOx sensor
feedbackfeedforward deltaP
Diesel Engine
O2 sensorPM sensor
Controller
HC added
•Separate devices shown for HC, NOx and PM control•Components depend on strategy:
With SCR, urea is addedWith LNT, HC is added in exhaust or by engine
•HC or external source necessary to heat PM filter for regeneration
18
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600 700
Catalyst Temperature (°C)
NOx
Conv
ersi
on (%
)
HC SCR
LNT
Urea SCR
Plasma
US06FTP
Typical Performance Curves of Slightly Aged Catalysts (Ford)• Widest temperature range for Urea SCR• Narrower temperature range for LNT which does not cover both FTP & US06• Plasma and HC SCR have lower NOx conversion• HC SCR conversion too low, but might improve with new materials
10
19
0102030405060708090
100
0.0% 2.0% 4.0% 6.0% 8.0% 10.0%FTP Fuel Economy Penalty
FTP
NO
x C
onve
rsio
n (%
)
LNT + filter
HC SCR + filter
Urea SCR + filter
Plasma + filter
Simulated FTP NOx Conversion vs Fuel-Economy Penalty Assuming 2.5L Engine (Ford)
• Lowest FE penalty for urea SCR, higher FE penalties for other strategies• Energy requirement for urea manufacture approximately 0.2%• FE penalty for LNT at 80% conversion exceeds technical target (<5%)
FE Penalty fromPM filter due toengine backpressure& filter regeneration
20
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8NOx Emission (g/mi)
PM E
mis
sion
(g/m
i)
Truck Passive NOxCar Passive NOxCar Active NOxTruck Active NOx
bin10
bin5
bin8
bin9
Bin 5-8 emission achieved for cars & trucks with active NOx systems, but durability is not proven
Emissions reduced with combustion improvements
System Testing with Fresh Catalysts Shows Progress to Bin 5Passive systems, which represent current production, do not achieve Bin 5Active systems studied in this program (LNT, urea SCR) are required for Bin 5Results are for fresh catalysts whose performance deteriorates with use
11
21
NOx Adsorber Aged for 1000 Hours (DOE APBF-DEC)
• Initial conversion met NOx target, but became inconsistent during aging• Some desulfations restore NOx conversion (even @ 1000hr), but others do not
• Remaining issues:• NOx conversion during aging frequently less than target• Process to desulfate, mitigation of fuel penalty, other regulated emissions (NMHC)
• Data guide other studies on mechanisms of catalyst deactivation (sulfur poisoning, thermal deactivation) and desulfation
• LNT aging studies are value added for OEM evaluation of this technology
• NOx adsorber for 1.9L engine aged using D15 fuel with 15-ppm sulfur
• Nine desulfations at 150hr, 300hr, then every 100hrs
• Test planned for an additional 1200hrs (120k mi equivalent)
• Test sequence designed for catalyst aging and not emission certification
• Conversion target equivalent to 0.07 g/mi NOx
Average NOx Conversion on FTP Schedule Simulated with Transient Engine DynamometerGaps in the line show desulfation events
40
60
80
100
0
200
400
600
800
1000
NO
x C
onve
rsio
n (%
)
TARGET
Time (hr)
22
Fast-Throughput Methods Used to Discover New, NOx Reducing Catalytic Materials (GM)
• Importance: Selective Catalytic Reduction (SCR) of NOx using HC has the potential to:- Simplify the emissions control strategy - Minimize fuel economy penalty - Avoid Infrastructure & dispensing issues
• Discovery of this new material is key to developing a HC SCR strategy for NOx reduction
• Status: Bench Reactor tests show that high temperature activity could be sufficient, but low temperature activity needs to be improved.
• Example of work with high risk/high potential reward
0102030405060708090
100
100 200 300 400 500 600Temperature(C)
NO
x Co
nver
sion
(%)
HC SCRNew Material
HC SCRExisting Material
NH3 SCR
0102030405060708090
100
100 200 300 400 500 600Temperature(C)
NO
x Co
nver
sion
(%)
HC SCRNew Material
HC SCRExisting Material
NH3 SCR
• Next Steps: Use the high throughput program to search the very large 3-5 component materials space to discover additional new materials that have good low temperature activity and higher high temperature activity. Move from Bench Reactor to engine exhaust testing.
12
23
Soot Regeneration using Microwave Spot Heating- Soot burning is initiated by localized microwave heating of less
than 5% of DPF volume.- Regeneration is less dependent on engine operation mode and is
controlled by the lower substrate and exhaust gas temperature.- Initiate soot regeneration at lower exhaust gas and DPF substrate
temperatures (more typical of diesel 200-250 o C)- Use un-catalyzed, Cordierite substrate - Minimizes fuel efficiency loss
Diesel Particulate Trap Regenerated by Microwave Spot Heating (GM)
Time, min
DPF Temp
o C
Potential of Uncontrolled Regeneration
1000
750
500
250
0
Current ExhaustHeating
Microwave Spot Heating
Time, min
DPF Temp
o C
Potential of Uncontrolled Regeneration
1000
750
500
250
0
Current ExhaustHeating
Microwave Spot Heating
Current Approach: Regeneration by Exhaust Heating- DPF peak temperature and soot burn rate is controlled by
limiting exhaust O2- Requires excess fuel in combustion chamber or in exhaust- Regeneration cycle is dependent on engine operating mode- Potential for uncontrolled temperature increase
Next Steps- Continue laboratory studies using
vehicle- loaded DPF substrates.- Transition to vehicle testing and
system integration.
Predicted MW Heating
Color shows MW field strength
45
7
Front
1 in.
2 in34
56Back
Entire DPF was Regenerated
Sections along length of DPF after heating front 5%
45
7
Front
1 in.
2 in34
56BackBack
Entire DPF was Regenerated
Sections along length of DPF after heating front 5%
24
• Importance– Kinetic maps summarize the response of new LNT materials to a standard test
protocol– The map data is used to calculate LNT performance for exhaust systems– Industry-wide consensus on LNT device maps simplifies supplier testing and
enables more rapid LNT development• Next steps
– Demonstrate and refine measurement protocol and then transfer the protocol to the aftertreatment suppliers
– Apply LNT experience to DPF and SCR* Crosscut Lean Exhaust Emissions Reduction Simulation
– CLEERS is a collaboration between global OEM’s, DOE, EPA, National Laboratories, Universities & TACOM to share non-competitive data and information
Response Maps for LNT Developed by CLEERS* ActivityNOx Concentration after LNTAbsorption 0-15min, Desorption 15-25 min
13
25
Enabling Technologies
Objective:• Develop sensors and measurement methods necessary for
implementation of control strategies
Barriers:• New combustion and emission control systems require new
sensors for system control and on-board diagnostic requirements
Strategy:• Sensors for NOx and particulates
• Particulate measurement during transients to improve engine control
• Sensors to control combustion
26
Particulate Sensor shows Potential for Engine Control (Honeywell, Univ of Minnesota)
• Importance– Goal is to develop a PM sensor that is suitable for closed-loop control of a
diesel engine and/or an on-board diagnostic monitor– The sensor requirements are low cost, robust to harsh environments and
manufacturable in high volumes. • Status
– Prototype sensors were fabricated and tested for more than 100 hours in the exhaust of different engines without degradation
– Sensor output correlates with traditional measurements and showsrequired temporal resolution
• Next steps– Continue testing, develop signal processing algorithms, correlate output to
individual cylinders
0
2
4
6
8
10
0 1 2 3 4 5 6Bosch Smoke NumberSe
nsor
Res
pons
e (v
olts
RM
S)
0
2
4
6
8
10
0 1 2 3 4 5 6Bosch Smoke NumberSe
nsor
Res
pons
e (v
olts
RM
S)
14
27
30 50 70 90 110 130 1500
10
20
30
40
50
Time (seconds)
LII (
ppb)
& R
PM/1
00
0
20
40
60
80
0
Spee
d (m
ph)
Real-Time PM Diagnostic Development (SNL) Real-time measurements of volatile fractionOn-road LII measurements
• Importance- Improved, real-time diagnostics are needed to investigate PM emissions
during transient operation and to meet the upcoming not-to-exceed regulations- Improved PM measurement sensitivity is needed for 2007 and beyond
• Status- Project enabled commercialization of laser-induced incandescence (LII) instrument
by Atrium Technologies, Inc.- Real-time volatile fraction measurement feasibility has been demonstrated
• Next steps- Develop volatile fraction diagnostic to enable possible commercialization- Develop real-time diagnostic for particulate size
Accelerationfrom a stop
340 360 380 400 420 440 460 480 500 520 540 560
0
1
2
3
4
Time (seconds)
Tota
l PM
–So
lid P
M(a
.u.)
0
20
40
60
80
100
120
Solid
Vol
ume
Frac
tion
(%)
EGRLoadRPM
28
CostObjective:
• $30/kW for engine, transmission and emission systems
Barriers: • Diesel engine costs are high relative to gasoline engines due to
added content • Cost optimization occurs during commercialization by an OEM
Strategy:• First, determine technical feasibility of powertrain and
aftertreatment system
• Then, improve the effectiveness of the system to reduce the costusing strategies such as LTC
15
29
Challenges and Needs
• Open Issues– Improvement of engine efficiency – Robust emission system– $30/kW cost goal
• Research Direction– More involvement of industry through vertically integrated
teams (auto, energy, suppliers) and new focused solicitations
– Continue work to improve understanding at national labs and universities
– Focus on new combustion regimes