Vehicle Simulation for Engine Calibration

to Enhance RDE Performance

Dr. Yutaka Murata

Yui Nishio

Dr. Yukihisa Yamaya

Masato Kikuchi

Honda R&D Co., Ltd.

Automotive R&D Center

1

IPG Apply & Innovate 2018

11st and 12nd of September, Karlsruhe, Germany

Background

Concept of model based simulation environment

Vehicle simulation model- Vehicle model

- Driver model

- Route model

- Traffic model

Application results in RDE-compliant powertrain development

Summary

Contents

2

New Civic diesel

Passed RDE regulation and achieved 91 g/km

91 g/km (6MT, Sedan)

93 g/km (6MT, Hachback)

109 g/km (9AT)

Fuel economy (CO2)

Exhaust emissions

Euro6d-TEMP

Modified NEDC

Civic

1.6L diesel engine

3

RDE definition

Difficulty to check RDE performance at all conditions during development

4

Chassis

dynamometer

RDE

Vehicle speed profile Fixed Depends on vehicle, driver, route, and traffic

Environment (Ta, Pa) Fixed Depends on season, weather, wind, and altitude

Road load forceStraight, w/o gradient

(w/o PEMS)

Depends on curves, altitude, road surface, passengers,

and baggage (with PEMS)

Repeatability with w/o

Method for RDE simulation and calibration

Necessity of model utilization for efficient development

5

Tests on road

Chassis dynamometer

(vehicle)

+ vehicle simulation

Engine test bed

(engine)

+ vehicle simulation

Model

(engine)

+ vehicle simulation

Vehicle simulation: consideration of road load force change

due to curves, altitude, road surface (weather, wind),

driver behavior, and traffic conditions

PEMS

Easy to simulate

and calibrateValidation in real world

EiL (Engine in the loop) MiL (Model in the loop)

Contents

6

Background

Concept of model based simulation environment

Vehicle simulation model- Vehicle model

- Driver model

- Route model

- Traffic model

Application results for RDE compliant powertrain development

Summary

Vehicle tests in (environmental) chassis dynamometer

Engine bench tests (EiL: Engine in the loop)

RDE-compliant engine development method

Engine calibration timeline

First

engine

First

vehicle

Vehicle model (virtual)

Target vehicle speed

- Fixed (NEDC, WLTC, etc) profile

- Vehicle simulation output profile

Model based simulation (MiL: Model in the loop)

Vehicle tests on road with PEMS

Engine

hardware

fix

SOPEngine

software

fix

Engine simulation environment

- ECU model

- Combustion model

- Catalyst model

Vehicle simulation environment

- Vehicle model

- Driver model

- Route model

- Traffic model

Validation phase

Vehicle simulation for engine hardware/calibration fix and RDE validation

7

Flowchart of model utilization

Coupling of vehicle simulation and engine simulation

8

Verification

Vehicle

test

Experiment

definition

Boundary

finderDynamic

DoE

Transient

measurement

(environmental

engine bench)

Dynamic

statistical

combustion

model

ECU

model

Catalyst

model

Vehicle simulation

(NEDC, WLTC, RDE)

Ne, Te, etc.

Simulation and

optimization

Synthetic

gas flow

test bed

Vehicle model

Driver model

Route model

Traffic model

Base maps, environmental

corrections, controllers,

aftertreatment control etc.

Calibration target

Engine simulation

Vehicle simulation

Predictive accuracy of engine model

Achievement of quantitative emissions prediction at RDE

9

NOx (mg/s)

Soot (mg/s)

gfuel (g/s)

Urban Rural

Motorway

Measurement

Simulation

Measurement

Simulation

Measurement

Simulation

Ve

hic

le

sp

ee

dN

Ox

(mg/s

)S

oo

t (m

g/s

)gfu

el(g

/s)

Engine

outlet

Tailpipe

Model input: engine speed, brake torque,

coolant temp., ambient temp., and ambient pressure

Contents

10

Background

Concept of model based simulation environment

Vehicle simulation model- Vehicle model

- Driver model

- Route model

- Traffic model

Application results in RDE-compliant powertrain development

Summary

Vehicle speed (km/h) Vehicle speed (km/h)

Co

astin

g tim

e (

se

c)

Ro

ad

lo

ad

fo

rce

(N

)

Road load force of vehicle model

Confirmation of the road load force accuracy by coasting simulation

Model input: Vehicle specification (F0, F1, F2)

11

Vehicle simulation

Vehicle simulation

Predictive accuracy of NEDC simulation

Measurement

Vehicle simulation

..

Target vehicle speed

Simulated vehicle speed

Upper limit

Lower limit

..

Engine speed and torque are well simulated (gear shift profile is given)

Measurement

Vehicle simulation

Model input: Vehicle specification (F0, F1, F2)

Target vehicle speed and gear shift profile

12

Predictive accuracy of WLTC simulation

Model input: Vehicle specification (F0, F1, F2)

Target vehicle speed profile

13

Target vehicle speed

Simulated vehicle speed

Upper limit

Lower limit

..

Engine speed and torque are well simulated (gear shift position is calculated)

Measurement

Vehicle simulation

..

Measurement

Vehicle simulation

Gear 1st->2nd

Gear 2nd->3rd

Gear 3rd->4th

Gear 4th->5th

Gear 5th->6th

Center

Vehicle speed distribution

Gear shift up timing distribution

Minimum

Center

Center

Maximum

Acceleration / deceleration distribution

Statistical analysis with big driving data

Extraction of driver behavior to reproduce real driving

14

Max. acceleration Max. shift up engine speed Max. acceleration &

shift up engine speed

15

Influence of driver model parameter on dynamic behavior

Succeed in targeting assumed driver characteristics

Too aggressive

Urban

Rural

Motorway

V*a pos: Vehicle speed multiplied by maximum positive acceleration

RPA: Relative positive acceleration

Too smooth

Urban

Rural Motorway

Route models for RDE performance validation

Introduction of detailed digital map information (traffic sign, curve, altitude)

16

n1

n2

n3 Simulation

Traffic flow model

Standard traffic

Heavy traffic

Simulation

Utilization of stochastic based traffic model to reproduce real driving

17

Standard traffic Heavy traffic

Contents

18

Background

Concept of model based simulation environment

Vehicle simulation model- Vehicle model

- Driver model

- Route model

- Traffic model

Application results in RDE-compliant powertrain development

Summary

Vehicle simulation for RDE route

Generation of vehicle speed by vehicle, driver, route, and traffic models

19

Digital map Simulation

Altitude

Urban

Rural

Motorway

Route

SimulationDigital map

Urban Rural MotorwayMotorwayUrban Rural

RDE measurement vs. simulation

Vehicle speed profiles at RDE conditions are well expressed in simulation

20

Simulation

Urban Rural

MotorwayReal driving

n1

n2

n3

Realization of vehicle simulation within RDE dynamic criteria regulation

21

Too aggressive

Too smoothUrbanRural

Motorway

Urban Rural

Motorway

Too aggressive

Too smoothUrban

Rural

Motorway

Urban

RuralMotorway

Real driving

Simulation

Real driving

Simulation

Driving characteristics at RDE

Simulation

9AT transmission model

6MT

9AT

22

Introduction of automatic transmission logic in the vehicle simulation

9AT Simulink

model

9AT6MT

Model based RDE performance evaluation

Achievement of emission prediction with vehicle and engine simulation

23Time (sec)

Engin

e s

peed (

rpm

)

To

rque (

Nm

)

Altitude (m

)

Vehic

le s

peed

(km

/h)Altitude

Vehicle speed

Tailpipe NOx

Gear shift position Brake torqueEngine speed

Vehicle, driver, route, and traffic models

Vehicle speed, engine speed, brake torque, and gear shift position

NOx, Soot, CO2 etc.

←Engine simulation

←Vehicle simulation

Urban

Rural

Motorway

Evaluation of emission robustness

Validity confirmation of hardware selection and calibration data settings

24

Total

(Urban+Rural+Motorway)

Urban

9AT

6MT

Realization of model based validation before actual measurement

Simulation

6MT Total

(Urban+Rural+Motorway)

Summary

Thank you very much for your kind attention.

25

It is a challenge to sufficiently validate RDE performance under all conditions through

road tests during vehicle development due to wide range of validating conditions.

A model based development technology was established to simulate, verify and

calibrate the emissions performance of a vehicle.

RDE performance could be accurately predicted by coupling a vehicle driving

simulation with an engine simulation that includes an ECU model, combustion model

(dynamic data based statistical model), and exhaust aftertreatment catalyst model.

Use of the simulation model enabled robust validation of RDE performance under

various conditions that assume driving on actual roads.

**

**

26