Modeling a PhlegmatizedDiesel-Engine in a Hybrid Electric ... · 3 Michael Auerbach, October 25th, 2010 Outline Motivation Introduction Phlegmatization DynamicSimulation EngineModel

Modeling a Phlegmatized Diesel-Engine in a Hybrid Electric Vehicle Using a Transient Predictive Model

Michael Auerbach, October 25th, 2010, Frankfurt a. M.

2 Michael Auerbach, October 25th, 2010

Institut für Verbrennungsmotoren und KraftfahrwesenUniversität StuttgartMichael Auerbach, Markus RufProf. Dr.-Ing. Michael Bargende

AUDI AGPre-Development Diesel EnginesDr rer. nat. René van DoornDipl.-Ing. Immanuel Kutschera


Outline

► Motivation

► Introduction

► Phlegmatization

► Dynamic Simulation

► Engine Model

► Overview

► Injection Modeling

► Transient Simulation

► Summary & Outlook


Motivation

Reduction of fleet

consumptionConservation of ressources

…

Diesel vehicles: High market share (EU)

Diesel Hybrid

Electric Vehicle

Public opinion

on emissions

►Main research aims for a P2-HEV-concept

Fuel consumption

Reduction of emissions

Driveability


torq

ue

engine speed

torq

ue

engine speed

IntroductionPhlegmatization

► General operating modes for Full-HEVs

► Electrical Driving

► Recuperation

► Load Shift of ICE

► Specific challenges for static Diesel-HEV emissions

NOXBSFC

high values

low values


IntroductionDynamic Effects on Emissions

► Effects on diesel engine emissions during dynamic load shift

► Inertia in air-system and control leads to miscalculation of cylinder state

► Decrease of load

� Lean zones in cylinder at high temperature

� Nitrous oxides

► Increase of load

� Rich zones in cylinder

� Soot

� High temperature � NOX in remaining lean zones

► Reducing dynamic of the ICE in the HEV

Phlegmatization of the Diesel Engine


IntroductionPhlegmatization

► Simulated phlegmatized ICE-operation during driving cycle

Electrical driving

Upshift in ICE load

Recuperation Downshift in ICE load

time [s]

500 510 520 530 540 550 560 570 580

torq

ue [

Nm

]

-160

-120

-80

-40

0

40

80

120

160

200

240 SOC vveh

TICE

MEM

velo

city [

km

/h]

0

10

20

30

40

50

60

70

80

SO

C [

-]

0.34

0.35

0.36

0.37

0.38

0.39

0.40


IntroductionTransient Simulations

► Why transient Simulation?

► Mean-Value models are fast but hide numerous disadvantages

� Steady-state look up tables do not reflect realistic behavior of air path and

heat up, therefore conditions for combustion may be incorrect

► Leaves the possibility to simulate emissions

� Delays in the air path and the control can be modeled

► Allows simulation of engine and exhaust aftertreatment warm-up

� Interaction with combustion model (wall heat transfer) and friction model

�Models of aftertreatment system can be added

► Inevitable for the simulation of effects on phlegmatization


Thermal Model

Aftertreatment Model

Engine Model*

EGR VNG

Engine Model Overview

► Requirements for a transient diesel engine model

► Various (fast) sub-models and appropriate controls

► Model scheme

► Injection Model for transient simulations

Soft - ECU

Injection

* 1D-Simulation and Combustion


Engine ModelInjection

► Measurements for each operating point (OP)

► High experimental effort

► Limited ability to simulate transient load/speed shifts

► Good and fast results at discrete OP’s

► Physical incompressible flow model

► High modeling effort

► Lot of data needed (geometries, coefficients, fluid properties,…)

► High calculation effort though high access to physical processes

► Empirical model

► Reasonable experimental effort

► Experimental data adequate for parameterization

► Accurate results combined with acceptable computational effort



inje

ction r

ate

[m

g/s

]

-505

10152025303540

time [s]

-0.0005 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.0030 0.0035

inje

cto

r p

ote

ntia

l [V

]

-200

20406080

100120140160

Measurements Preprocessing Model Generating Integration

200 µs @ 1600 bar 500 µs @ 1600 bar 800 µs @ 1600 bar 1100 µs @ 1600 bar

► Fingerprint measurements

► Gradually incrementing injection duration at constant rail pressure level

► Repeating measurement for various rail pressure levels

[mgs]


inje

ctio

n r

ate

[m

g/m

s]

-5

0

5

10

15

20

25

30

35

40

time [s]

-0.0005 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.0030 0.0035


► Data preprocessing

► Extending gradient of rising and falling edge

► Defining start of injection (SOI) and end of injection (EOI)


SOI EOI



► Data preprocessing

► Extending gradient of rising and falling edge

► Defining start of injection (SOI) and end of injection (EOI)

► Saving profiles to database

Measurements Preprocessing Model Generating Integrationin

jectio

n r

ate

[m

g/m

s]

-5

0

5

10

15

20

25

30

35

40

time [s]

-0.0005 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.0030 0.0035


rail

pre

ssure

[bar]

0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

injection activation duration [ms]

0 100 200 300 400 500 600 700 800 900 1000 1100


► Using RLTDependenceProfXYZ

► Used RLTs are rail pressure and injector activation duration

► Profiles from database are processed as XYTables in a XYZMapOfTables





InjMultiProfileConn

SOAAD

pRail

SOI

ID

∆pdrop

Soft - ECU

∆dmInj/dt

► Start of Injection

► Variable Profile Dependency Object

► RLTs (ID, pInj)

►Mass flow dropmInj

∆tInj


injectiondelay



► From Soft ECU: Start of activation, activation duration, rail pressure

Injectionrate [mg/s]

time [s]

TDCSOI(MI)SOI(PI1)SOI(PI2)

SOA(MI)

Crank Angle [°CA]


inje

ction

ra

te [

mg

/°C

A]

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

crank angle [°CA]

-50 -30 -10 10 30

crank angle [°CA]

-50 -30 -10 10 30

time [s]0 5 10 15 20 25 30

IME

P [

bar]

0

4

8

12

16

20

24

28

inje

cte

d m

ass [

mg]

0

10

20

30

40

50

60

70

80

Transient SimulationResults

► Validation: Load shift @ 2000 rpm

► Time-scale results

� Engine model vs. Engine test bench

► Cuts at discrete time steps (°CA-based)

� Injector model vs. Injector test bench

measurement simulation


Summary & Outlook

► Diesel-HEV is important for future mobility

► Control-strategy must focus on emissions, fuel consumption and driveability

► Main effect on emissions is Phlegmatization

► Transient engine simulation is inevitable for surveys on this effect

► Fast injection model and control needed for transient simulations

► Approach to a phenomenological injection model was shown

► Outlook

► Coupling of further models (thermal, exhaust aftertreatment,…)

► Using model for pre-application and testing of Diesel-HEV-Strategy


Thank you for your attention!

Acknowledgements

Thanks for support and inspiration to

Michael Grill (FKFS),

Björn Lumpp (MAN) and

Torsten Rausch (AUDI AG)

Modeling a PhlegmatizedDiesel-Engine in a Hybrid Electric ... · 3 Michael Auerbach, October 25th, 2010 Outline Motivation Introduction Phlegmatization DynamicSimulation EngineModel

Documents