Multicylinder Engine Control Using TurboSupercharging: Modeling and Experiments

1
Multicylinder Engine Control Using
Turbo/Supercharging Modeling and Experiments
Hunter Mack and Greg Bogin Professors J-Y Chen
and R.W. Dibble University of California -
Berkeley
LTC University Consortium Meeting Sandia National
Laboratory Feb 7-8, 2007
2
The UC Berkeley Roadmap(with speed bumps
included)
Install New FTM Controls
Engine Failure 1 (May 2006)
Install Supercharger
Engine Failure 2 (Dec 2006)
Skeletal Mechanisms
New Ion Sensor Software
Ion Modeling
Test FTM Supercharger
3
Broken valve ingested and imbedded into the
engine head
4
Two pistons destroyed, cylinder walls damaged by
broken valve
5
The engine has been repaired and is now running
again
6
Outline

• Further Development of Isoctane Reduced/Skeletal
  Mechanisms (Prof Chen, Yuk Fai Tham, Fabrizio
  Bisetti)
• HCCI Multi-cylinder Operation
• Actuation (Andrew Armey)
• Control Strategies (Nick Killingsworth, Shuo
  Yang)
• Supercharged HCCI (Andrew Armey, Robert Mills)
• HCCI and Ion-Sensor Technology (Hunter Mack,
  Gregory Bogin)

7
Development of Isooctane Skeletal Mechanism
8
Modeling RoadmapDevelopment of Isooctane
Mechanisms
Detailed
857
( of species)
Skeletal
295
(better emissions)
258
215
(algorithm 2)
(algorithm 1)
Reduced
63
Reduced w/ QSSALTC Search Algorithm
9
Development of Skeletal (215 species) and Reduced
(63 Species) mechanisms
Based on LLNL Iso-octane Mechanism (857)

• Detailed Mechanism (857 species)? Skeletal
  Mechanism (215 species) by removing unimportant
  species and steps
• Skeletal Mechanism ? Reduced Mechanism with QSSA
  (63 species)
• Comprehensive Validation of Reduced 63
• - autoignition
• - transient HCCI single zone WMR with emissions
• - premixed flames
• - Kiva3v

10
Identification of Species in QSS (1 of 2)
Species in Quasi-Steady State ? WpWd

• Estimated error based on production destruction
  rates
• Estimated error based on Cqss(1)
• Time scales based on CSP (Computer Singular
  Perturbation) (2,3)
• Above approaches are found not adequate with
  isooctane mechanism with complex pathways

• Turanyi et al On the error of the qssa J. Phys.
  Chem, 1993,97,163-172
• Massia et al An algorithm for the construction
  of global mechanisms with CSP data
• Combust Flame, 1999, 117, 685-708
• Lu et al Complex CSP for chemistry reduction and
  analysis Combust Flame, 2001, 126,
• 1445-1455

11
Identification of Species in QSS (2 of 2)
Species in Quasi-Steady State ? WpWd

• Estimated error based on production destruction
  rates
• Estimated error based on Cqss(1)
• Time scales based on CSP (Computer Singular
  Perturbation) (2,3)
• Above approaches are found not adequate with
  isooctane mechanism with complex pathways
• ? New approach Automatic search guided by
  targeted flames

12
CSP time scale analysis leads to a reduced
mechanism that fails ignition when species with
times lt 3E-4 ms are not included
Isooctane T800K, P 40atm, ?0.3 Constant
Pressure Reactor
Reduced mechanisms lt 145 species fail to predict
ignition

• Errors induced
• by QSSA are not
• accounted for
• Nonlinear
• interactions among
• QSS species

13
Chemical Pathways During Autoignition
T gt1200K HO2? OOH T1000K
H2O2M?2OHM T lt850K RO2M?RO2M
RO2?QOOH QOOH?QOOH QOOHO2 ? O2QOOH
O2QOOH?OQOOHOH OQOOH?OQO OH
2nd ignition 1000K
1st ignition
14
Targeted Conditions to Represent HCCI Engine
Operation States
WMR P-T relations
1
With combustion
near walls
2
3
3
4
5
2
6
4
motoring
5
6
1
1000K
6 conditions
CR16 without boost pressure
LTC
HCCI engine conditions Near TDC
High temperature
15
Comparison of Autoignition Delays
Skeletal 215 vs detailed
Reduced 63 vs detailed
16
Comparison of HCCI Single-Zone WMR
Reduced 63 species vs detailed 857 species
Burning with EGR
Bimodal burning
VW TDI geometry CR16, 1800 RPM
17
Emissions of HCCI Single-Zone WMR
HC- unburned hydrocarbons OHC-Oxygenated
unburned hydrocarbons
18
Validation of Reduced 63 Species Mechanism using
Kiva3v
Pressure data from LLNL
S.M. Aceves, J. Martinez-Frias, D. Flower, J.R.
Smith, R.W. Dibble, and J.-Y. Chen SAE Paper
2002-01-2870 (2002).
19
Reduced Mechanism based on single targeted
condition has limited application range
(LTC includes three regimes of auto-ignition)
target
target
Applicable regime
Applicable regime
Target 700K 10 atm inaccurate at high
temperatures
Target 1000K 40 atm inaccurate in LTC
20
HCCI Control Strategies
21
Summary of Accomplishments in Control of HCCI
Combustion Timing

• Initially demonstrated two methods (electrical
  trim heaters and exhaust throttles) for
  controlling HCCI combustion timing
• Built real-time HCCI engine controller hardware
• Discovered low cost alternative
  start-of-combustion (SOC) sensors a) ion sensor
  and b) microphone sensor, essential for HCCI
  engine control.

22
Control requires an actuation method to change
combustion timing
Version 1 Electrical Trim Heaters

• Fast heat heaters coupled with mixing control
  of hot and cold air to control intake
  temperatures.
• Independent control of cylinder temperature by
  using separate intake runners.
• Effective method for achieving control of
  ignition time, engine balancing.

23
Control requires an actuation method to change
combustion timing
Version 2 Exhaust Flap Valves

• Exhaust throttles are used to adjust the
  combustion timing
• Closing throttle increases in -cylinder
• residual gases for next cycle and advances
  combustion timing
• Emulates variable valve timing (VVT)
• Sub-optimal, but rapidly implemented

24
Control requires an actuation method to change
combustion timing
Version 3 Fast Thermal Management (FTM) using
Intake Flap Valves
25
Improved Butterfly Valve / Supercharger Control
System
Version 4 Intake Valve redesign using Hot Air
Throttling
The system uses a supercharger, with one air line
going through a heater and one going through an
intercooler.
26
The flow was modeled in the mixing T
30 degrees open
45 degrees open

• Used COSMOS FloWorks to determine average
  temperature at outlet of valve system.
• Assumed that the air moving through the heater
  would have higher pressure, because of the
  pressure drop across the intercooler.
• Input conditions were 293 K at 101 kPa for the
  low temp. line, and 393 K at 103 kPa for high
  temp. line

27
Computed Temperatures Show Good Controllability
no change after 60 degrees
90 degrees
0 degrees
28
The Butterfly Valve system is installed and
connected to new Controller
Valves controlled by DC motors on a H-Bridge
cicuit
29
Multicylinder Control Strategy
Previous Multicylinder control strategy
intelligence Custom Simulink/MATLAB system Jason
Souder (PhD) _at_ eControlsParag Mehresh (PhD) _at_
Caterpillar
Develop improved/expanded control system using NI
hardware (Nick Killingsworth LLNL/UCSD) (Shuo
Yang UC Berkeley)
30
Simple model used to derive feedforward controller
Feedback control accounts for errors
Ref Killingsworth Flowers (LLNL)
31
Proof of Concept LLNL CAT3406 Experimental Data
using the same controller design
Ref Killingsworth Flowers (LLNL)
32
Control /DAQ Setup(with support from National
Instruments)
Ion Sensor Set-up
Desktop Computer
DAQ 3
NI cRioChassis
DAQ 1
DAQ 2
Counter/Timer 1
Optical Encoder
WoodwardSmartFlameModule
H-bridge
Ion Sensors
Electric Motors
Valve Position Sensors
Pressure Transducers
33
Super/Turbocharged HCCI
34
The upper limit of the HCCI operating Regime can
be extended through Turbocharging
35
An increase in Boost Pressure advances combustion
timing
36
Hardware sized, obtained, and installed

• Uses Eaton M-45 roots-types supercharger, SpearCo
  air-to-water intercooler, and custom manifold and
  butterfly Tee valves.
• Belt sizing for supercharger

37
Supercharger with Custom Manifold
38
Key Questions we want to answer.
How does an increase in Intake Pressure affect
Combustion Timing
Power Output
Ion Signal
Cross-talk between cylinders
39
Testing the Supercharger a Preliminary Success
Slight Boost Test
Next Step Sizing the Pully on the Supercharger
(Data From Tuesday Feb 5)
Pressure (bar)
Shows that engine rebuild successful and new DAQ
set-up is working
CAD
40
Ion Sensors for HCCI
41
Ions in flames and Ion-sensor technology
Chemi-ionization reaction is the principal source
of ions in hydrocarbon flames
This reaction is rapidly followed by the charge
exchange reaction
H3O is the dominant ion in both fuel lean and
slightly rich hydrocarbon flames
42
The ionization mechanism focuses on the formation
and consumption of H3O .
The ion mechanism have 34 reactions involving 9
ionic species (CHO,O,N,OH,H3O, NO, O2,
N2, electrons)
43
We modified a 4 cylinder 1.9 Liter diesel engine
to use in HCCI mode.
VW TDI Engine specifications
Intake Manifold
VW TDI 1.9 Liter in HCCI mode
44
Experiments show that CAD for 50 cumulative heat
release is directly correlated to 50 of ion
signal.
45
The ion signal accurately detects the combustion
event
Experimental Results Propane fuel
46
Using Acetylene increases the ion signal in an
HCCI engine
Remember the initiation reaction for ionization
Acetylene produces large number of CH radicals.
Electrical Aspects of Flames by Weinberg, 1969.
Ion signal should be stronger for acetylene
compared to propane.
47

Experimental Results Acetylene fuel
48
Numerical modeling of ion production in Propane
49
N-Heptane shows promise in ion production under
lean conditions
F0.32
50
Conclusions

• A 63 species reduced mechanism for isooctane
  provides accurate (and more computationally
  efficient) predictions for HCCI
• Engine control can be achieved through Fast
  Thermal Management(New Hardware Controller in
  testing phase)
• Supercharging can extend HCCI operating
  range(Testing in progress, Preliminary Results
  look good)
• Our Ion Mechanism numerical model results are in
  good agreement with previous experimental
  findings(New software in testing phase)