Validation and Verification of ANSYS Internal Combustion Engine Software. Martin Kuntz, ANSYS, Inc.

Transcription

1 Validation and Verification of ANSYS Internal Combustion Engine Software Martin Kuntz, ANSYS, Inc.

2 Contents Definitions Internal Combustion Engines Demonstration example Validation & verification Spray box Combustion Port flow applications IC engine applications Wednesday, October 10, Automotive Simulation World Congress 2

3 Contents Definitions Internal Combustion Engines Demonstration example Validation & verification Spray box Combustion Port flow applications IC engine applications Wednesday, October 10, Automotive Simulation World Congress 3

4 Definitions Verification Verify, that model is implemented correctly Characteristics Simplified geometry Focused on single physical model Compare to analytical or other CFD Validation Demonstrate simulation accuracy Characteristics Realistic geometry A combination of physical models Compare to experimental data Demonstration Illustrate application of software to generic case Characteristics Realistic geometry A combination of physical models No comparison to data Wednesday, October 10, Automotive Simulation World Congress 4

5 Contents Definitions Internal Combustion Engines Demonstration example Validation & verification Spray box Combustion Port flow applications IC engine applications Wednesday, October 10, Automotive Simulation World Congress 5

6 IC Engine Simulations Types Component simulations Intake port, intake manifold, water jackets, fuel injectors Spray bomb IC engine simulations Cold flow Charge motion Combustion Thermal management Emissions Wednesday, October 10, Automotive Simulation World Congress 6

7 ICE Simulation Workflow Internal combustion engine simulation components Preprocessing Geometry decomposition Initial meshing Simulation parameter definition Moving deforming meshes Smoothing, remeshing, layering Particle tracking Injection, tracking, evaporation, wall-interaction Combustion Ignition, flame front propagation Post-processing Automatic processing of monitor and solution data Wednesday, October 10, Automotive Simulation World Congress 7

8 Contents Definitions Internal Combustion Engines Demonstration example Validation & verification Spray box Combustion Port flow applications IC engine applications Wednesday, October 10, Automotive Simulation World Congress 8

9 Demonstration : Direct Injection Gasoline Engine Complete cycle setup Initial conditions and boundary conditions provided by 1D simulation Material Iso-octane Spray injection 6-hole injector Double injection Transient mass flow Prescribed diameter distribution Liquid evaporation model Spark ignition G equation combustion Testcase provided by BMW Courtesy: BMW Wednesday, October 10, Automotive Simulation World Congress 9

10 Demonstration : Direct Injection Gasoline Engine Initialization Burned conditions at EVO Boundary condition Specified temperature Mesh size: cell count (TDC) to (BDC) Wednesday, October 10, Automotive Simulation World Congress 10

11 Demonstration : Direct Injection Gasoline Engine Simulations Cold flow run Charge motion Plus spray injection Plus particle tracking Combustion Plus ignition Plus flame front propagation Wednesday, October 10, Automotive Simulation World Congress 11

12 Demonstration : Direct Injection Gasoline Engine Cold flow simulation Results for CFX and Fluent Cylinder averaged values of pressure and temperature Wednesday, October 10,

13 Demonstration : Direct Injection Gasoline Engine Cold flow simulation Results for CFX and Fluent Swirl ratio, tumble ratio Wednesday, October 10,

14 Demonstration : Direct Injection Gasoline Engine Velocity vector plots CA 330 CA 440 CA 555 CA 715 Wednesday, October 10, Automotive Simulation World Congress 14

15 Demonstration : Direct Injection Gasoline Engine First injection Wednesday, October 10, Automotive Simulation World Congress 15

16 Demonstration : Direct Injection Gasoline Engine Evaporation Wednesday, October 10, Automotive Simulation World Congress 16

17 Demonstration : Direct Injection Gasoline Engine Second injection Wednesday, October 10, Automotive Simulation World Congress 17

18 Demonstration : Direct Injection Gasoline Engine Total particle mass Influence of wall film model

19 Demonstration : Direct Injection Gasoline Engine Combustion simulation Wednesday, October 10, Automotive Simulation World Congress 19

20 Demonstration : Direct Injection Gasoline Engine Burned flow simulation Cylinder averaged values Pressure Temperature Mixture fraction Wednesday, October 10, Automotive Simulation World Congress 20

21 Contents Definitions Internal Combustion Engines Demonstration example Validation & verification Spray box Combustion Port flow applications IC engine applications Wednesday, October 10, Automotive Simulation World Congress 21

22 Validation: Particle Tracking Particle injection Primary / secondary breakup Injection type: cone / hollow cone Tracking Particle-wall interaction Wall film modeling Evaporation Testcases Bosch spray box (cold spray) Hiroyasu spray box (cold spray) Koss spray (hot spray)

23 Validation: Bosch Spray Box Two cases based on experimental data from ROBERT BOSCH GmbH Details of experimental setup Gas parameters Setup # 1 Setup # 2 Gas type N 2 Temperature [K] 300 Pressure [MPa] Fuel Properties Fuel type Heptane Density [kg/m 3 ] Surface tension [kg/s 2 ] Spray parameters Initial temperature [K] 300 Nozzle diameter [mm] Injection pressure [MPa] 10 Injection velocity [m/s] 138 Particle mass flow rate [g/s] 1.5 Injection rate, single pulse [ms] 1.5 Estimated initial spray angle [deg] Injection Weber number m sampling point (0, , 0.03) Available data Spray penetration over time m Sampling point (0, , 0.03) - Droplet diameter distribution - Droplet velocity distribution Kumzerova, E. and Esch, T., Extension and Validation of the CAB Droplet Breakup Model to a Wide Weber Number Range, Proc. of the 22nd Europ. Conf. on Liquid Atomization and Spray Systems, Paper ILASS08-A132, Como Lake, 2008.

24 Validation: Bosch Spray Box Mesh dependence study Grid Number of cells Size of the cell near the nozzle [m 2 ] (radial x axial length) Coarsest e-4 x 12.8e-4 Coarse e-4 x 6.4e-4 Medium e-4 x 3.2e-4 Fine e-4 x 1.6e-4 Finest e-5 x 8e-5 Coarsest Coarse Medium Fine Finest Wednesday, October 10, Automotive Simulation World Congress 24

25 Validation: Bosch Spray Box Mass penetration Mesh dependence study Fluent simulation Setup #2 Wednesday, October 10, Automotive Simulation World Congress 25

26 Validation: Bosch Spray Box Local spray results at sampling point Droplet diameter distribution Local velocity distributions KH-RT breakup model Setup #1

27 Validation: Bosch Spray Box Local spray results at sampling point Droplet diameter distribution Local velocity distributions KH-RT breakup model Setup #2

28 Validation: Bosch Spray Box Mass penetration Comparison KH-RT and SSD break-up model Fluent simulation Setup #2 Wednesday, October 10, Automotive Simulation World Congress 28

29 P e n e t r a t i o n D e p t h [ m ] Validation: Bosch Spray Box Mass penetration Comparison of break-up models CFX simulation Setup # E x p e r i m e n t N o b r e a k u p R e i t z & D i w a k a r S c h m e h l T A B E T A B C A B M e d i u m g r i d ( 4000 n o d e s ) d t = 2 e - 6 s T i m e [ s ]

30 Validation: Hiroyasu Spray Box Gas parameters Case 1 Case 2 Case 3 Gas type N 2 Temperature [K] Pressure [MPa] Fuel Properties Fuel type C 12 H 26 Density [kg/m 3 ] 840 Surface tension [kg/s 2 ] Spray parameters Initial Temperature [K] Injection Velocity [m/s] Particle Mass Flow Rate [g/s] Nozzle diameter [mm] Injection rate, single pulse, ms

31 Validation: Hiroyasu Spray Box Mass penetration Case 1, 2 and 3 Fluent simulation

32 Validation: Koss Spray Box Gas Temperature [K] 800 Gas Pressure [MPa] 5 Gas Type N 2 Particle Mass Flow Rate [g/s] 4.62 Droplets type nheptane (C7H16) Evaporating spray Liquid penetration at 90% spray mass fraction Density [kg/m 3 ] 684 Surface tenstion [N/m 2 ] 0.02 Nozzle diameter [mm] 0.2 Injection rate [ms] 1.3 Droplet diameter [mm] 0.2 Injection Velocity [m/s] 215 Initial spray angle [deg] 10.1 Liquid penetration Length Measurements: H. Koss, D. Bruuggemann, A. Wiartalla, H. Backer, and A. Breuer, Results from Fuel/Air Ratio Measurements in an N-Heptane Injection Spray, IDEA periodic report, RWTH Aachen, 1992.

33 Validation: Koss Spray Box Breakup model comparison TAB ETAB CAB Reitz CFX run Wednesday, October 10, 2012

34 Validation: Koss Spray Box Comparison Fluent KH-RT model CFX TAB model Wednesday, October 10, 2012

35 Contents Definitions Internal Combustion Engines Demonstration example Validation & verification Spray box Combustion Port flow applications IC engine applications Wednesday, October 10, Automotive Simulation World Congress 35

36 Validation: Hamamoto Testcase Premixed combustion in a closed vessel with fixed wall Propane/air mixture: Equivalence ratio =1.0 Measured data Optical access Pressure transducer Described in Hamamoto et al. (1988) Ewald (2005) Wednesday, October 10, Automotive Simulation World Congress 36

37 Validation: Hamamoto Testcase Average cylinder pressure Mesh sensitivity study Fluent, 2d quad meshes Comparison CFX vs. Fluent

38 Validation: Hamamoto Testcase Average cylinder pressure Comparison of mesh size and types 3D hexahedral and tetrahedral meshes

39 Combustion Validation: Pancake Engine Premixed combustion Flat head, flat piston, SI ICE Fuel: C 3 H 8 References: Alkidas (1980) Han and Reitz (1997) Simulated interval [-30;30 CA] ATDC Piston motion Displacement [m 3 ] 0.82 x 10-3 Bore x Stroke [mm] x Compression ratio 8.56 Connecting rod length [mm] 158 TDC clearance [mm] 12.6 Equivalence ratio 0.87 Engine speed [rpm] 1500 Spark timing [deg. ATDC] -27 Volumetric efficiency 40 Wednesday, October 10, Automotive Simulation World Congress 39

40 Combustion Validation: Pancake Engine Average cylinder pressure G equation combustion Comparison CFX Fluent Wednesday, October 10, Automotive Simulation World Congress 40

41 Contents Definitions Internal Combustion Engines Demonstration example Validation & verification Spray box Combustion Port flow applications IC engine applications Wednesday, October 10, Automotive Simulation World Congress 41

42 ICE Validation IC engine applications Public engine cases Collaborations with customers Benchmark for customers Of interest Valuable experimental data No confidentiality No restrictions for publication Wednesday, October 10, Automotive Simulation World Congress 42

43 ICE Validation: Engine Applications Port flow simulation Thobois generic engine Cold flow simulation Bosch: direction injection diesel engine with PIV data Partially premixed combustion Wisconsin: direct-injection spark-ignition engine Premixed combustion Ducati: premixed engine setup Diesel combustion Engine cooling simulation Wednesday, October 10, Automotive Simulation World Congress 43

44 ICE Validation: Port Flow Steady intake flow conditions Testcase defined in Large Eddy Simulations in IC Engine Geometries Thobois, Rymer, Souleres, Poinsot SAE Paper, Port length 132 mm Inner port diameter 16 mm Outer port diameter 34 mm Valve opening 10 mm Cylinder length 300 mm Cylinder diameter 120 mm Wednesday, October 10, Automotive Simulation World Congress 44

45 ICE Validation: Port Flow Comparison RANS SST model LES SAS model CFX results 2 2 Isosurface S 10 Color eddy viscosity Wednesday, October 10, Automotive Simulation World Congress 45 6

46 ICE Validation: Port Flow 20 mm Mean Axial Velocity Axial Velocity Fluctuation

47 ICE Validation: Port Flow 70 mm Mean Axial Velocity Axial Velocity Fluctuation

48 Validation: Bosch Engine Experiment Investigation of Diesel engine in-cylinder flow with Particle Image Velocimetry (PIV) and High-Speed PIV Generation of a comprehensive and high-quality database for Large Eddy Simulation Simulation RANS & Scale resolving turbulence models, e.g. LES, DES CFX simulation Publications A Strategy for Evaluation of LES Applied to Diesel Engine In- Cylinder Flow Joint Effort of Simulation and Experimental PIV Flow Analysis Les Rencontres Scientifiques de l'ifp LES for Internal Combustion Engine Flows November 2010 Analysis of In-Cylinder Air Motion in a Fully Optically Accessible 2V-Diesel Engine by Means of Conventional and Time Resolved PIV 9TH INTERNATIONAL SYMPOSIUM ON PARTICLE IMAGE VELOCIMETRY PIV 11, Kobe, Japan, July 21-23, 2011

49 Swirl Ratio [-] Validation: Bosch Engine RANS simulation 5 4 Flow characteristics swirl R s z 2 N V V r v 2 r dv dv Simulation Average swirl on planes 0 Experiment (plane data) CFX Swirl ICE ration Swirl in cylinder [cad] Wednesday, October 10, Automotive Simulation World Congress 49

50 Validation: Bosch Engine Scale Resolving Models LES Large Eddy Simulation DES Detached Eddy Simulation SAS Scale Adaptive Simulation Grid size: nodes Wednesday, October 10, Automotive Simulation World Congress 50

51 Validation: Bosch Engine Comparison of velocity profiles Sample line: -10mm below dome Exp: 2D-2C absolute velocity Sim: 3D-2C absolute velocity Sample Line

52 Validation: Wisconsin Engine Research project conducted at University of Wisconsin sponsored by ANSYS Inc. Characterization of Direct-Injection Spark-Ignition Operation and Investigation of Particulate Matter Formation" Research work of single-cylinder direct-injection spark-ignition engine November 2011 November 2013 Wednesday, October 10, Automotive Simulation World Congress 52

53 Validation: Wisconsin Engine Riccardo Hydra Base Modern DISI engine architecture Engine Specifications Engine Type Chamber Geometry Fueling type 4-Stroke, 4-Valve, SI Pentroof Spray-guided directinjection Single-cylinder direction-injection spark-ignition engine used for measurements Displacement cm 3 Compression Ratio 12:1 Injection Pressure 11 MPa Wednesday, October 10, Automotive Simulation World Congress 53

54 Validation: Wisconsin Engine Phase 1 single cylinder metal engine Motored operation Fired homogeneous (fully vaporized) spark-ignition operation with premixed air/fuel mixtures Direct-injection spark-ignition operation Phase 2 detailed investigations Detailed spray characterization measurements in a spray vessel Laser-based in-cylinder measurements to characterize the velocity field or fuel distribution Detailed measurements of particulate matter number count Wednesday, October 10, Automotive Simulation World Congress 54

55 Cylinder Pressure [bar] Validation: Wisconsin Engine Motored Engine Measurements Repeatability 2 Influence of coolant temperature Volume [L] In-cylinder pressure for 2000 rpm 80 kpa, 80 o C intake, 80 o C coolant Wednesday, October 10, Automotive Simulation World Congress 55

56 Validation: Ducati Engine 4-stroke S.I. P.F.I. race engine Premixed combustion Validation / verification Pressure trace Collaboration University of Bologna, Ducati Motor Holding & ANSYS Publication Flexible Meshing Process and Multi-cycle Methodology for Simulating Reacting Flows in High Performance SI Engines with ANSYS CFX International Multidimensional Engine Modeling User s Group Meeting 2010 (IMEM 2010), Detroit, April 12 th, 2010 Wednesday, October 10, Automotive Simulation World Congress 56

57 Validation: Ducati Engine Simulation with CFX Efficient multi-cycle methodology Single-cycle initialization / multi-cycle initialization Influence of mesh resolution & types Wednesday, October 10, Automotive Simulation World Congress 57

58 Average cylinder pressure [bar] EXP. MEAN IN-CYLINDER PRESSURE [bar] Validation: Ducati Engine Simulation with Fluent Variation of combustion models / turbulent flame speed models and settings rpm - Cyl_Horiz rpm - Cyl_Vert ECFM - cycle 1 C-eqn - cycle 1 G-eqn - cycle 1 G-eqn zcd - cycle 1 G-eqn p - cycle 1 G-eqn pc - cycle Wednesday, October 10, CRANK ANGLE [deg. ATDC] Crank Angle [deg] 2012 Automotive Simulation World Congress 58

59 Validation: Engine Cooling Simulation Perform Engine Cooling Simulation Requires simultaneous simulation IC engine simulation Cylinder head coolant channel simulation Thermal inertia of two models is orders of magnitude different two separate simulations Wednesday, October 10, Automotive Simulation World Congress 59

60 Validation: Engine Cooling Simulation Steady state conjugate heat transfer of cylinder head Transient combustion simulation in diesel engine in-cylinder Time-averaged Heat Flux profile on cylinder head Iterative Process Temperature profile on the firedeck

61 Validation: Engine Cooling Simulation Import: temperature data Iteration process Export: time-averaged heat flux from IVC to EVO

62 Validation: Engine Cooling Simulation Heat Flux n-heptane 1 step mechanism Effect of mesh resolution y+~ 200 y+~ 30 y+~ 20 y+~ 1 Wednesday, October 10, Automotive Simulation World Congress 62

63 Validation: Engine Cooling Simulation Heat Flux n-heptane 1 step mechanism Effect of combustion and turbulence model K-epsilon Laminar Finite Rate SST K-w K-epsilon Finite Rate - Eddy SST K-w Wednesday, October 10, 2012 Laminar Finite Rate Finite Rate - Eddy 2012 Automotive Simulation World Congress 63

64 Validation: Engine Cooling Simulation Temperature data at locations of thermocouples Heat Flux n-heptane 1 step mechanism + SST k model (Laminar Finite Rate) Wednesday, October 10, Automotive Simulation World Congress 64

65 Summary Validation and verification examples related to internal combustion engines Basic spray and combustion cases Port flow applications Cold flow IC engine applications Combustion IC engine applications Good agreement of results for most cases in different application areas Ongoing work in the ICE validation project at ANSYS Continuation of work with existing engines Collection of new validation cases Reference cases for current ICE software and future software developments Wednesday, October 10, Automotive Simulation World Congress 65

66 Any Questions? Wednesday, October 10, Automotive Simulation World Congress 66