Modeling Constant Volume Chamber Combustion at Diesel Engine Condition

Transcription

1 Modeling Constant Volume Chamber Combustion at Diesel Engine Condition Z. Hu, R.Cracknell*, L.M.T. Somers Combustion Technology Department of Mechanical Engineering Eindhoven University of Technology *Shell Technology Center Thornton

2 Objective Fundamental study on AUTOIGNITION due to FUEL and CONDITION. Study THEIR effects on MIXING AND COMBUSTION. TOOL: Experimental (Constant Volume Vessel) Numerical (Computational Fluid Dynamics) PAGE 1

3 Experiment Combustion Research Unit (CRU) Constant volume vessel provided with Bosch 6 hole nozzle injector n-heptane (CN~54) and iso-octane (CN~15) are used to study fuel effects Chamber volume Inner diameter Inner length Injector details Technical details L 100 mm Min 70 mm Chamber temperature K Bosch 6 hole nozzle Chamber pressure Injection pressure (P i ) 2-75 bar bar PAGE 2

4 Determination of Injection Period and Rate Fuel volume controlled by injector pressure P i and injection period. Reference test done at P i =1600bar, injection period =1.9ms. Idealized profile for modeling showing on the right, when injecting n-heptane and isooctane at Pi=900bar, injection period =1.3ms. Solid: n-heptane; dash: iso-octane PAGE 3

5 n-heptane injection and autoignition (pressure rise v.s. time) T=863K P=50bar PAGE 4

6 iso-octane injection and autoignition (pressure rise v.s. time) T=863K P=50bar PAGE 5

7 Mathematical Model Chemical Kinetics: 137 species and 633 reactions for n-heptane/isooctane/toluene * J.C.G.Andrae, T.Brinck and G.T.Kalghatgi, HCCI experiments with toluene reference fuels modelled by a semidetailed chemical kinetic model, Combustion and Flame 155 (2008) Combustion Modeling: Flamelet Generated Manifolds (FGM) * Reduced Chemistry Method; Instead of solving hundreds of species equations, only limited PDE are solved for reacting flows. PAGE 6

8 Flamelet Generated Manifolds (FGM): Flamelet Generated Manifolds (FGM) Concept: process described by Z,Y only 2 variables: mixture fraction: *1,2 denotes fuel and oxidizer stream in reacting flows reaction progress variable: Y Y M CO2 CO CO *speicies dominant in the products stream is chosen to represent reaction progress variable 2 Y M CO Y M CH 2 CH O 2 O PAGE 7

9 Flamelet Generated Manifolds (FGM): Implication: Conservation Equations: PAGE 8 Z j Z j j j x Z D x x Z u t Z Y j Y j j j x Y D x x Y u t Y Fuel injection FGM pre-processing Also for information such as temperature, species mass, etc. * + mass conservation, momentum conservation and ideal gas equation Fuel Oxygen Products Z Y Mixture Fraction Reaction Progress Variable

10 FGM pre-processing pre-processing Flamelet calculation (diffusion) tabulation FGM f ( Z, Y) Manifolds Dominant for diesel combustion Homogeneous reactor; Livengood-Wu, decide ignition Premixed flamelets Flame Index? Diffusion flamelets PDF integration FGM ~ ~ ''2 f ( Z, Y, Z, Y ''2 ) PAGE 9

11 CFD Modeling Solver: Star-CD Spray model: Reitz-Diwakar/Reitz model built in STAR * C.Bekdemir, E.P.Rijk, L.M.T. Somers, L.P.H de Goey, B.A. Albrecht, On the Application of the Flamelet Generated Manifod (FGM) Approach to the Simulation of an Igniting Diesel Spray, SAE Technical Papers, Turbulence: high-reynolds k RANS to account for turbulence from fuel injection, initial air in chamber is quiescent. Combustion: FGM integrated into Star-CD via user-subroutine. PAGE 10

12 CFD Model Mesh: r, z, ) (50@1mm,62@1mm,180@2 o ) Time step: during injection and combustion Injection: n-heptane / isooctane into quiescent high T, high P air corresponding to experimental cases Injection profile: PAGE 3 s injector PAGE 11

13 n-heptane injection (pressure trace history, solid-experiment) T=803K; P=30bar T=863K; P=30bar PCCI? Combustion starts after injection is finished, premixed but not totally premixed. PAGE 12

14 n-heptane injection: T=803K, P=30bar (simulation results: temperature at EOI) PAGE 13

15 n-heptane injection: T=863K, P=30bar (simulation results: temperature at EOI) PAGE 14

16 n-heptane injection (mixture distribution at EOI; stoichiometric-grey zone) T=803K; P=30bar T=863K; P=30bar PAGE 15

17 iso-octane injection: T=863K, P=50bar (Left: pressure trace, experiment: dots, simulation: line ) Temperature snapshot at the time (red line) indicates possible ignition location PAGE 16

18 Conclusions & Future Work Experimental work shows the effect of fuels and conditions on autoignition and combustion. Modeling work shows capability of capturing features of mixing and combustion due to different conditions and fuels. Future work includes further identification on initiation of autoignition, both experimentally (lasers, schilieren photography, etc.) and numerically (Livengood-Wu integral, etc.); chemical kinetics for rich combustion should also be addressed. PAGE 17

19 Acknowledgements Acknowledgements to R.Cracknell, B.Somers, D.Bradley and G.Kalghatgi. Acknowledgements to A.Mcdougall, T.Davies, etc for work on CRU, as well as to many other colleagues in Fuels and EVT group in Shell. Acknowledgements to students and faculties in TU/e for many assistances. PAGE 18

20 Keyword Analysis Autoignition is important for diesel engines. Fuel autoignition property affects the engine performance. IGNITION FUEL CONDITION (T,p) PERFORMANCE MIXING COMBUSTION PAGE 19