Question? Leave a message!




Chemical Kinetics, HCCI & SI Combustion

Chemical Kinetics, HCCI & SI Combustion 15
Part 3: Chemical Kinetics, HCCI & SI Combustion Reciprocating Internal Combustion Engines Prof. Rolf D. Reitz Engine Research Center University of Wisconsin-Madison 2014 Princeton-CEFRC Summer School on Combustion Course Length: 15 hrs (Mon.- Fri., June 23 – 27, 2014) Copyright ©2014 by Rolf D. Reitz. This material is not to be sold, reproduced or distributed without prior written permission of the owner, Rolf D. Reitz. 1 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Short course outline: Engine fundamentals and performance metrics, computer modeling supported by in-depth understanding of fundamental engine processes and detailed experiments in engine design optimization. Day 1 (Engine fundamentals) Part 1: IC Engine Review, 0, 1 and 3-D modeling Part 2: Turbochargers, Engine Performance Metrics Day 2 (Combustion Modeling) Part 3: Chemical Kinetics, HCCI & SI Combustion Part 4: Heat transfer, NOx and Soot Emissions Day 3 (Spray Modeling) Part 5: Atomization, Drop Breakup/Coalescence Part 6: Drop Drag/Wall Impinge/Vaporization/Sprays Day 4 (Engine Optimization) Part 7: Diesel combustion and SI knock modeling Part 8: Optimization and Low Temperature Combustion Day 5 (Applications and the Future) Part 9: Fuels, After-treatment and Controls Part 10: Vehicle Applications, Future of IC Engines 2 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion http://www.erc.wisc.edu/combustion.php Modes of engine combustion HCCI uses a hybrid combustion strategy. Premixed fuel and air is inducted, but instead of igniting with a spark as in a SI engine, the high temperature from compression causes the mixture to spontaneously react, like in a diesel engine. Ignition occurs at slightly different times at different locations in the chamber. One feature of HCCI combustion is how quickly the fuel is consumed. 3 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Curtis, 2014 Chemistry: importance of fuels 4 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Daw, 2013 New advanced combustion regimes HCCI 5 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Basic combustion concepts – Spark Ignition (SI) How can SI engines operate with engine speeds from 100 to 20,000 rev/min? 2 T Kinetic energy, kV piston burned Integral length scale l L I piston x Turbulence Kinetic energy dissipation rate, 3 e V /L piston piston S T fuel/air 2 Diffusivity, D k /e V L piston piston Because turbulent flame speed, S , scales with rpm T Characteristic Time Combustion (CTC) model Reitz & Bracco, 1983; Abraham, 1985 Species conversion rate (Y, species mass fraction, local equilibrium solution) i t k/e L / V ; c piston piston dY i V Mallard-Le Chatelier propagating wave speed: SD  piston T dt Glassman, 1996 6 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Halstead, 1977 Basic combustion concepts – Diesel (CI) Shell Ignition Model Q Ignition  R RH +O 2R 2 Delay  B R R + P + Heat  R R + B  R R + Q Af04  R + Q R + B  B 2R  R termination  2R termination Switch to Characteristic Time Combustion model Turbulence generated by fuel injection t k/e L / V nozzle nozzle c Kong, 1992 7 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Turbulent mixing Hot products with Cold reactants Spark-ignition t k/e burned unburned L / V piston piston High turbulence S T - faster combustion Diesel Injected fuel with entrained air air fuel t k/e L / V air nozzle nozzle Delayed ignition (PCCI) - better mixing S =0 T Matalon, 2011 8 CEFRC2-3, 2014 S /S T L Part 3: Chemical Kinetics, HCCI & SI Combustion Summary of combustion regimes • Gasoline engine spark-ignition with flame propagation: High turbulence for high flame speed  heat losses. Issues: NOx and UHC/CO, knock (CR, fuels), throttling losses  low thermal efficiency TE 25% • Diesel engine with spray (diffusion) combustion: Rich mixtures (soot) & high temperatures (NOx)  higher TE 45% • H/Premixed Charge Compression Ignition – LTC, chemistry controlled (CR): Sensitive to fuel, poor combustion/load control, low NOx-soot  TE 50% spark-ignition diesel H/PCCI 9 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Williams, 1988 Premixed volumetric combustion & chemical kinetics Species and energy conservation equations  cs ii  (uD )   ( )    i i i t  Constant volume combustion – Well-Stirred-Reactor (WSR) n r dY W '' ' ii  (  ) (Y,T), i 1,...,n  k,, i k i k s dt  k1 ' ''  k,, i k i nn ss I specific internal energy      YY ii  (Y ,T)       k f,, k b k chemical label WW ii  11 ii    n reactions r n species s ; reactant/product stoichiometric coefficients mass fraction n s  dT 1 e() T dY ii molecular weight (YY ,TT ) ( , )  dt c (Y,T) W dt i1 vi e species energy i e i 10 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Law, 2006 Homogeneous charge: no spatial gradients Y i  / 7 i A 1.610 (cm,mol,s), b 1.83, E 11.6(kJ / mol)  t 13 1.4 10 0 2 n s h  T f ,i i    13 1.2 10 tc i1 p 1.5 13 k 1 10  Consider single overall reaction 12 8 10 CH 4 k 1 12 k 6 10 CH OH  CH  H O 4 3 2 12 4 10 0.5 b k AT exp( E / RT) 12 2 10 0 0  d CH CH 4 4 0 500 1000 1500 2000 2500  kCH OH  4 dt W T CH 4 11 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Reitz, 1981 0.6 m=0.25 HCCI: Ignition delay FU() 0.5 Y i 0.4  / i m=0.5  t 0.3 0 n s h  T f ,i i Cold boundary 0.2  m=1.0  tc difficulty i1 p 0.1 U 0 Consider single component system 0 0.2 0.4 0.6 0.8 1 U U TT  unburned U  1.0 TT  burned unburned Example: dU mm 1  F(U)  bU (1U) dt Ignition 11 delay : mt b For U  U o mm UU 0 bt U So, time to reach, say, 5U : 0 o 4 Ignition delay: 1 bt  m mbt  Const. F (m,m;(1 m);U) 21 m 5mU 0 U 12 CEFRC2-3, 2014 m+1 m F(U)=U (1-U)Part 3: Chemical Kinetics, HCCI & SI Combustion Warnatz, 2006 Combustion chemistry models – CH (15 spec, 31 react.) 4 CH + 2 O = CO + 2 H O 4 2 2 2 Methyl Reactions Hydrogen-Oxygen Chain 13 CH + O  CH O + H 3 2 1 H + O  OH +O 14 CH +OH  CH O + H +H 2 3 2 2 H + O  OH + H 15 CH +OH  CH O + H 2 3 2 2 c 3 H + OH H O + H 16 CH + H  CH 2 2 3 4 4 H O + O 2 OH 23 CH + H  CH + H 2 3 2 2 Hydroperoxyl Formation and Consumption 28 CH + OH CH + H O 3 2 2 b 5 H + O + M  HO + M Formaldehyde Reactions 2 2 6 HO + H  2 OH 17 CH O + H CHO + H 2 2 2 7 HO + H  H + O 18 CH O + OH CHO + H O 2 2 2 2 2 8 HO + H  H O + O Formyl Reactions 2 2 9 HO + OH  O + H O 19 CHO + H CO + H 2 2 2 2 Conversion of Carbon Monoxide to Carbon Dioxide 20 CHO + OH CO + H O 2 End 10 CO + OH  CO + H 21 CHO + O  CO + HO 2 2 2 Methane Consumption 22 CHO + M CO + H + M Initiation Start 11 CH + H  H + CH 4 2 3 H atom abstraction 12 CH + OH  H O + CH Methylene Reactions 4 2 3 24 CH + O  CO + H 2 2 2 2 Methylidyne Reactions 25 CH + O  CO + OH + H 2 2 27 CH + O  CHO + O 2 26 CH +H  CH + H 2 2 31 CH + OH  CH O + H 2 29 CH + OH  CH O + H 2 2 30 CH + OH  CH + H O 2 2 Conversion to products by sequential fragmentation by H abstraction 13 CEFRC2-3, 2014 H O chemistry 2 2 High temperature Part 3: Chemical Kinetics, HCCI & SI Combustion Lu, 2009 Brakora, 2013 Chemical kinetic mechanisms for engine simulations Requirements for mechanisms for practical engine simulations: • Size can not be too large due to CPU time limitation 100 species • Capable of predicting auto-ignition delay time accurately • Contain proper reactions for pollutant formation precursors Biodiesel surrogates - Significant mechanism reduction is required. Soy biodiesel - Methyl: - palmitate (C16:0) - stearate (C18:0) - oleate (C18:1) - linoleate (C18:2) CH 4 - linolenate (C18:3) C C C 2 4 9 14 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Warnatz, 2006 Hydrocarbon kinetics - NTC Second Stage Ignition H O = OH + OH 2 2 Acceleration by 1400 Q•OOH branching 1200 O 2 1000 . 800 600 400 200 0 555 580 605 630 655 680 705 730 Initial Temperature (K) First Stage Ignition Isomerization steps 15 CEFRC2-3, 2014 Induction Period (s)Part 3: Chemical Kinetics, HCCI & SI Combustion Warnatz, 2006 Alkane fuel oxidation Example: Propane Energy release CO + OH CO2 + H High Smaller R HCO temperature 1200K Chain branching CH , C H ,… 3 2 5 reaction b-scission H + O OH + O 2 Blue flame 1100K Thermal decomposition Fuel: RH R H O OH + OH +OH, H,… 2 2 + n-C H • + O 3 7 2 Negative Internal H atom abstraction Chain propagation temperature 800 QOOH ROO 900K coefficient QO + OH +O 2 Olefin channel OOQOOH Internal H atom Chain branching abstraction Cool flames 800K HOOQ’O + OH HOOQ’OOH •CH CH CH OOH 2 2 2 OQ’O + OH Isomerization 16 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Mehl, 2009 HCCI combustion kinetics Typical HCCI Combustion H R• Temperature and Heat Release Rate Fast High Temperature profiles - RH Combustion • • + O 2 OO• H O 2 2 + HO • 2 OOH • O + + •OH + O 2 O + •OH OOH Ethers/ olefins •OO Degenerate - •OH Branching Path O O CAD TDC •OH O + + HOO • Aldehydes/ketones 17 CEFRC2-3, 2014 HRR T, P Part 3: Chemical Kinetics, HCCI & SI Combustion Patel, 2004 Mechanism reduction – identify key reaction steps ERC n-heptane mechanism Burnout stage 6 6 6 1 1. n . n- -C C H H + + O OH H= = C C H H - -2 2+ + H H O O 7 7 16 16 7 7 15 15 2 2 2 2. C . C k ke ett = = C C H H C CO O + + C CH H O O + + O OH H 7 7 12 12 5 5 11 11 2 2 nd 2 st 3 3. H . H O O + + M M = = O OH H + + O OH H + + M M stage 2 2 2 2 1 stage 4 4.. H HO O + + H HO O = = H H O O + + O O 2 2 2 2 2 2 2 2 2 2 5 5.. C CH H + + H HO O = = C CH H + + H H O O 4 4 2 2 3 3 2 2 2 2 2 6 6.. C CO O + + O OH H= = C CO O + + H H 2 2 3,4,5 , 7 , 7 7 7. C . C H H O O + + O O = = C C k ke ett + + O OH H 7 7 15 15 2 2 2 2 7 7 12 12 1,7 Time 18 CEFRC2-3, 2014 Temperature Part 3: Chemical Kinetics, HCCI & SI Combustion Ra, 2008 Reduced mechanisms: match shock tube and RCM data First stage (t1), main ignition (t2=tig) delay 10 Expts: Fieweger, 1997 1 0.1 Cal, tig Exp, tig Predicted ignition delay times Cal, t1 validated against shock tube tests Exp, t1 (data from Fieweger) 0.01 =1.0 and P=40 bar n-heptane/air 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1000/initial temperature 1/K 19 CEFRC2-3, 2014 ignition delay msPart 3: Chemical Kinetics, HCCI & SI Combustion Ra, 2011 Mechanism reduction methodology tetra-decane: ROO+O =R-keto+OH 2 10 Reduction of reaction pathways and species number baseline A = k×A base k=2 - combination of chemical lumping, graphical k=0.5 1 reaction flow analysis and elimination methods Reaction rate optimization - ignition delay curve sensitivity analysis 0.1 Pre-exponential: A = k×A base 0.01 600 700 800 900 1000 1100 1200 1300 1400 Ignition delay sensitivity coefficient initial temperature K (logtt  log ) 10kk 10 12 ignition delay sensitivity ST ( ) 100 25 60 ig gradient sensitivity log t log (k k ) 10 base 10 1 2 50 20 40 Ignition delay gradient sensitivity coefficient 30 15 20 d log t d log t 10kk 10 12 10 10  dT dT 0 ST ( )100 gr 5 log (kk ) -10 10 1 2 -20 0 -30 Positive S : counter-clockwise rotation gr -5 -40 Negative S : clockwise rotation gr 650 750 850 950 1050 1150 1250 1350 1450 initial temperature K 20 CEFRC2-3, 2014 ignition sensitivity % ignition delay ms gradient sensitivity coefficientPart 3: Chemical Kinetics, HCCI & SI Combustion Ra, 2011 Mechanism reduction – group reaction classes Ignition delay sensitivity coefficient No Reaction A B C A B C effect s s s r r r I RH  H  R  H O 2 II RH  OH  R  H O ●↓ ●↑ ●C τ , p , p , τ 1 2 3 4 2 III RH  HO  R  H O ○↓ ○↓ ○↓ τ , P , P , τ 1 2 3 4 2 2 2 IV RH  O  R  HO ○↑ ○↓ ● P , P 2 3 22 V R O ROO 2 VI-a ROO = QOOH ●↓ ●↓ ○C τ , P 1 2 VI-b QOOH + O = OOQOOH ○↓ ●↓ P , P 2 3 2 VI-c ○↓ ●↓ P , P OOQOOH = R-keto + OH 2 3 VII R-keto = CH O + R'CO + OH ●↓ ○C τ 1 2 VIII R'CO = X + X + CO 12 R = S + S + S IX ●↑ ○ P , P , τ 2 3 4 1 2 3 Ignition delay gradient Sensitivity of ignition delay curves of n-heptane oxidation sensitivity coefficient - solid circle, open circle and blank entry denote dominant, mild and not significant influence, respectively. - C indicates counter-clockwise rotation. - Circle only indicates clockwise rotation. 21 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Ra, 2008 ERC-MultiChem: PRF 41 species, 158 reactions  base mechanism Source mechanisms: LLNL n-heptane 100 (560 species; 2,539 reactions), isooctane =1.0, 40 bar (857 species; 3,606 reactions), 10 ERC n-heptane (29 species; 52 reactions) iC8 10 PRF90 PRF80 PRF 1 PRF60 MultiChem nC7 Exp, Fieweger et al. (1997) Exp, iC8 1 0.1 Exp, PRF90 Exp, PRF80 Exp, PRF60 Exp, nC7 0.1 100 0.01 PRF nC7H16/air 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 MultiChem 1000/initial temperature 1/K Exp, Fieweger et al. (1997) 10 0.01 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1000/T 1/K 1 0.1 =1.0, iC H /air, 40 bar 8 18 0.01 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1000/T 1/K 22 CEFRC2-3, 2014 ignition delay ms ignition delay ms ignition delay msPart 3: Chemical Kinetics, HCCI & SI Combustion Ra, 2011 Chemical class grouping: “MultiChem” skeletal mechanism Physical property Chemistry LLNL Detailed ERC reduced surrogates surrogates mechanism mechanism Cyclo alkanes cyclohexane cyclohexane decalin n-alkane n-dodecane n-heptane n-alkane n-tetradecane n-octadecane iso-alkanes heptamethyl nonane 857 species 25 species iso-octane tetramethyl hexane 3586 reactions 51 reactions Aromatics naphthalene mcymene toluene tetralin n-pentylbenzene n-heptylbenzene 23 CEFRC2-3, 2014 MultiChem Mechanism 100 species, 348 reactions Part 3: Chemical Kinetics, HCCI & SI Combustion Ra, 2011 8 Surrogate fuels: n-heptane, Ignition delay validations - “MultiChem” iso-octane, tetradecane, cyclohexane, toluene, decalin, ethanol, MB/D…… Gauthier CNF 2004 Fieweger CNF 1997 10 100 1000 Model Model Experiment Exp, Fieweger et al. (1997) Experiment Model 100 10 1 10 Propane iC8H18 1 phi=1.0 0.1 phi=1.0 nHeptane 1 Pin= 30 bar Pini=40 bar phi=1.0 0.1 Pini=40 bar Fieweger CNF 1997 0.1 0.01 0.9 1 1.1 1.2 1.3 1.4 1.5 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 0.01 1000/T 1/K 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1000/T 1/K 1000/T 1/K EXP (Bounaceur et al.) 10000 10000 10 Bounaceur et al. Experiment Andrae et al. Model 1000 ERC-MultiChem 1000 1 100 MCH Decalin Phi=1.0 phi=1.0 100 0.1 10 Pini=40 bar Toluene Experiment Model 1 10 0.01 0.55 0.6 0.65 0.7 0.75 0.8 0.8 0.9 1 1.1 1.2 1.3 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1000/T 1/K 1000/T 1/K 1000/T 1/K Bounaceur IJCK 2005; Andrae CNF 2005 Shen,Energy & Fuels, 2009 24 CEFRC2-3, 2014 ignition delay micr-sec ignition delay ms tig n mic ro-s ignition delay ms ignition delay ms ignition delay msPart 3: Chemical Kinetics, HCCI & SI Combustion Amsden, 1997 3-D CFD modeling Solve conservation equations on (moving) numerical mesh Mass Species Momentum combustion source terms Energy 25 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Perini, 2014 3-D CFD: Improved solver numerics LLNL n-heptane mech. Jacobian structure ERC PRF mechanism Jacobian structure LLNL MD mechanism Jacobian structure Sparse analytical Jacobian formulation Sparsity of hydrocarbon fuel mechanisms increases n s with size non-zeroes: 860; sparsity: 62.7% non-zeroes: 3571; sparsity: 86.2% non-zeroes = 49763; sparsity: 99.7% 47 (62.7%) 160 (86.2%) 2878 (99.7%) SpeedCHEM performance scaling 2 i 10 Y  / i  n s  t 1 10 LLNL nC H 0 n 7 16 s ERC h  T f ,i i multichem  0  10 ERC tc i1 p PRF LLNL LLNL -1 MD PRF 10 All functions and equations are evaluated in function, y' = f(y) -2 matrix form 10 Jacobian, J(y) = df(y)/dy ERC nC H linear system solution 7 16 ODE system function, analytical Jacobian -3 10 evaluation and linear system solution 1 2 3 4 10 10 10 10 achieve linear scaling with n number of species, n s s 26 CEFRC2-3, 2014 time per evaluation msPart 3: Chemical Kinetics, HCCI & SI Combustion Perini, 2014 3-D CFD: Improved solver numerics SpeedCHEM ignition delay time calculation scaling 4 10 9 reaction mechanisms tested Direct dense Jacobian 3 10 SpeedCHEM, direct sparse -n = 29 to 7171 s SpeedCHEM, Krylov 2 3 -n = 52 to 31669 10 r  n s 1 10 18 ignition delay calculations per 0 mech 10 -1 10 - phi = 0.5, 1.0, 2.0  n s - T = 650, 800, 1000 K -2 0 10 - p = 20, 50 bar 0 -3 - t = 0.1 s 10 1 2 3 4 10 10 10 10 number of species Promising, efficient approach for practical engine simulations 1.Numerically exact solution (no mechanism reduction or manipulation). 2.Speed-up of more than three orders of magnitude at large sizes (n 1000) s 3.Even for modest sizes (50-500 species), overall CPU time for chemistry is reduced by 3-10 times in comparison with dense chemistry integrators 4.Preconditioned Krylov solution for future, very large mechanisms 27 CEFRC2-3, 2014 CPU time sPart 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2009 Shi, 2012 Efficient chemistry solvers – cell clustering Perini, 2014 Group thermodynamically-similar cells to reduce the calling frequency to save computer time - Adaptive Mechanism Clustering (AMC) scheme Extended dynamic adaptive chemistry (EDAC) scheme Dynamically determine the size of fuel chemical mechanism based on the local and instantaneous thermal conditions of the cells Thermodynamically similar cells (similar temperature, equivalence ratio ) Chemkin Solver Remap back to cells 28 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Shi, 2012 HCCI engine validation Full AMC AMC+EDAC ERC PRF mech. (39 sp, 141 rxn) 48.27 hrs. 3.99 hrs. 2.88 hrs. 3.0 2.5 2.0 1.5 1.0 Experiment 0.5 Simulation-Full Chemistry Simulation-AMC + EDAC model 0.0 -50 -40 -30 -20 -10 0 10 20 30 40 50 Crank Angle 29 29 CEFRC2-3, 2014 Pressure (MPa)Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2006 SI engine combustion modeling Flame propagation Models with Detailed Chemistry Turbulent Flame Propagation • G-equation description of combustion Spark plug • Laminar and turbulent flame speeds • Primary heat release calculation • Flame quench due to mixture stratification Post-flame Chemistry • CO oxidation, H -O reactions 2 2 • Pollutant formation mechanisms Knocking Combustion • Auto-ignition mechanisms • Location / intensity Spark Ignition Engine 30 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Law, 2014 What is a turbulent flame? A T S T A  L Ensemble of thin (laminar) flamelets, interacting with the flow turbulence. Due to increased surface area, turbulent flame “brush” propagates at enhanced velocity A  L SS  TL A T 31 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Turbulent flame structure Combustion regime diagram Kolmogorov/Batchelor Peters, 2000 length scales: Quenched broken 1/ 4 3   3/ 4 reaction zones ll  Re KI  e  ll  Engine flames K  Flamelets ll  KF Ghandhi, 2012 Laminar flame thickness: 0 / c If l C l C 0.1l , local S = 0   pT K m33  m F T 0 l  20 mm F  S Liang, 2007 uL It is not possible to resolve a turbulent flame on a practical engine simulation grid 32 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Reitz, 1981 Laminar flame speed: balance between reaction and diffusion 2 YY ii  D  / i 2 tx dY i - Mallard, Le Chatelier S = D TT 2 N TT dt 0  D  h  / c  f ,i i p 2 tx i1 Consider the single component system: 2  UU TT  mm 1 unburned  D F() U and F(U) bU (1 U) with U  2  tx TT  burned unburned 1.0  Admits a traveling wave solution mS () xSt 1/ m D U(x  St)  1/(1 e ) S U where Db D D(m 1) and  S  2 0 mS bm m 1 x 33 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Farrell, 2005 Laminar flame speed C-C-OH C-C-C C=C-C methanol 0.90 1-pentene ethanol propene 1-butene 1-heptene 0.85 benzene 1-hexene 1-octene ethane n-heptane anisole 3-heptene 0.80 cyclohexane cyclopentane 2-butene 2-pentene ethylbenzene i-propylbenzene n-hexane cyclopentene n-butane 0.75 t-butylbenzene propane n-propylbenzene n-pentane methylcyclopentane 2-methyl-2-butene MTBE 0.70 2-methyl-1-butene toluene iso-octane iso-octene iso-pentane 0.65 iso-butane o,p xylene neopentane methane 0.60 1,2,4-trimethylbenzene m-xylene 0.55 1,3,5-trimethylbenzene Paraffins Olefins Aromatics Oxygenates 34 CEFRC2-3, 2014 Peak Burning Velocity (m/s)Part 3: Chemical Kinetics, HCCI & SI Combustion Lutz, 1988 Importance of chemistry - Methane Flux analysis: oxidation proceeds through methyl – slow path, high activation energy due to tight C-H bonds 3 Rich conditions Acetylene 35 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Lutz, 1988 Importance of chemistry - Ethane • Limited flux through slow methyl channels – hydrogen abstraction leaves weaker secondary C-H bonds 2 • Greater flux through chain branching pathways, e.g., H + O  OH + O 2 • Ignition delays are very sensitive to rates of H atom production Acetylene Olefin Ignition delay Ethane Ignition delay Methane 36 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Wang, 2012 Diffusion: Turbulence models Production (RANS - RNG k-e P  u u S i j ij u ’ + U = u u i i i i Mean flow strain rate Reynolds stresses 2 k = 3u /2 i l = U t i t t  k/e t turbulent/mean flow time scale 2 D  C k /e T m 37 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 Ignition and level set (G-equation) models Partially Premixed Flame (DI Engine) Discrete particle ignition model Φ 1 O , O, NO … 2 Diffusion G-Equation Flame CO , H O, CO, NO… Φ ≈ 1 2 2 propagation Burnt Gas Diffusion End Gas CH , CO, H, H … 4 2 Φ 1 Auto-ignition (detailed kinetics) Fuel Droplets S from flame speed correlations T End-gas Flame Post-flame Burned gas: G0 Zone Front Zone Unburned gas: G0 G  0 u (v v )G  S G  D k G f vertex T T t  38 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 Turbulent flame speed correlations 1/ 2 1/ 2 2  22    C  t t     S a b l a b l u l m2 ign 2 4 3 4 3 T    1 1 exp     ab   43  S  I t 22 b l b l S l   L 0 1 F 1 F l F       Progress Term Peters, 2000 3/ 2 1/ 2 Discrete Particle    l u l    F F u Stretch factor:   I 1  2    Ignition Kernel 0   15l S r     L  K K (DPIK) model Turbulence stretch Curvature k t  Characteristic Timescale: e 1.5 k r r  C l  C 0.16 Transition criterion: k k m11 I m e  G  Fan, 2000 0 u (v v )G  S G  D k G f vertex T T t  39 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 Laminar flame speed correlations Power Law (Metghalchi & Keck, 1982):  b 30     T p Metghalchi et al.     S  S 1 2.1Y  L L,0 dil     Present study T p Liang et al.  0   0  25 20   2.18 0.8( 1) 15 b  0.16  0.22( 1) 10 Reference State: 300K, 1bar 2 S  B  B (  ) 5 L,0 m 2 m Liang et al. : 0 0.0 0.5 1.0 1.5 2.0 2.5  2 S   exp (  )   L,0 Equivalence Ratio,    -0.134 For iso-octane,       26.9 3.86 1.146 40 CEFRC2-3, 2014 0 S (cm/sec) L,refPart 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 Validation - PFI/DI gasoline engines Based on MIT PRF Mechanism Bore × Stroke 89 mm × 79.5 mm (25 species, 51 reactions) Compression Ratio 12 : 1 Model constants: Cm1=2.0, Cm2=1.0 Engine Speed 1500 rev/min (Fixed in all cases) PFI Mode Spark timings (ATDC) -44, -40, -36, -32 MAP (kPa) 65 DI Mode (Spark timing sweeps) Spark timings (ATDC) -32, -28, -24, -20 MAP (kPa) 75 End of Injection (ATDC) - 72 DI Mode (Manifold-Absolute-Pressure sweeps) MAP (kPa) 75, 80, 90, 100 Spark timing (ATDC) - 33 End of Injection (ATDC) - 68 DI Mode (End-Of-Injection sweeps) End of Injection (ATDC) -76, -72, -68, -64 MAP (kPa) 75 DI Configuration Spark timing (ATDC) - 32 41 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 Validation - PFI engine operation Spark Timing = 40 BTDC CA = -20 ATDC CA = -5 ATDC CA = 10 ATDC CA = 20 ATDC Evolution of the G=0 surface Evolution of Temperature 42 CEFRC2-3, 2014 Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 Validation - PFI engine operation 2.5 2.5 EXPT EXPT Spark Timing SIMU SIMU 2.0 2.0 PFI mode PFI mode -44, -40, -36, O O -32 ATDC -36 ATDC 1.5 1.5 -32 ATDC Engine Speed 1.0 1.0 1500 rev/min 0.5 0.5 0.0 0.0 -100 -50 0 50 100 -100 -50 0 50 100 o o Crank Angle ( ATDC) Crank Angle ( ATDC) 2.5 2.5 EXPT EXPT SIMU SIMU 2.0 2.0 PFI mode PFI mode O O -44 ATDC -40 ATDC 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0 -100 -50 0 50 100 -100 -50 0 50 100 o o Crank Angle ( ATDC) Crank Angle ( ATDC) 43 CEFRC2-3, 2014 Pressure (MPa) Pressure (MPa) Pressure (MPa) Pressure (MPa)Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 Role of flame propagation Explore Kinetics-Controlled Formulation for Turbulent Flame Propagation: After ignition kernel stage, each cell is modeled as a WSR, detailed chemistry is applied. “Flame propagation” is controlled by heat conduction and auto-ignition. 2.5 3.5 EXPT EXPT 3.0 G-equation G-equation 2.0 Kinetics only Kinetics only 2.5 DI mode PFI mode 1.5 2.0 Spark timing 1.5 1.0 Spark timing -32 ATDC -44 ATDC 1.0 0.5 Transition from kernel to G-eqn 0.5 Transition from kernel -20 ATDC to G-eqn at -20 ATDC 0.0 0.0 -80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80 o o Crank Angle ( ATDC) Crank Angle ( ATDC) dY i SD  Mallard-Le Chatelier propagating wave speed: T dt 44 CEFRC2-3, 2014 Pressure (MPa) Pressure (MPa)Part 3: Chemical Kinetics, HCCI & SI Combustion Liang, 2007 G-equation Kinetics Controlled Role of flame propagation PFI case Spark timing = -44 ATDC Summary: Auto-ignition chemistry alone is NOT sufficient to properly model flame propagation. Turbulence enhancing effect on flame propagation speed in SI engines CANNOT be neglected. 45 CEFRC2-3, 2014
Website URL
Comment