Co-Simulation for hybrid vehicle control software development · Co-Simulation For HEV Control...

Co-Simulation For HEV Control SW-Development

Co-Simulation for hybrid vehicle control software development

Marcus Boumans, Sebastian WanslebenRobert Bosch GmbH

GT User Conference21. October 2013

Frankfurt

DGS-EC/EAM | 06/09/2013 | © Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing, distribution, as well as in the event of applications for industrial property rights.

Diesel Gasoline Systems

11


ContentsContents Scope & Motivation

Use cases & Requirements

Simulation Environments

Parameterization & Validation

Summary and Conclusion



2


Scope of (Co)Simulation

Scope & Motivation

Scope of (Co)SimulationSupport the development of innovative

energy management control strategies for passenger cars

Domains under investigation:

Combined control strategy

HEV Powertrain control strategyThermal system control strategy



3


Motivation for (Co)Simulation

Scope & Motivation

Motivation for (Co)Simulation Good reproducibility and fast repeatability of experiments

Required especially for traffic simulation & warm-up experiments

Investigation of real world systems with limited availability Not all HEV / thermal topologies are available as real systems

Time & costs savings by “Virtual Controller Prototyping” Run control strategy model on a rapid prototyping system in a

vehicle

Using (Co)Simulation makes us faster and saves money!



Using (Co)Simulation makes us faster and saves money!

4


Contents

Usecases & Requirements

Contents Scope & Motivation







5


Development: Thermal System Control Strategy


Development: Thermal System Control StrategyFunction: Optimized thermal management with

i iti ti f f t d f l prioritization of comfort and fuel consumption

Considered information:Considered information: Engine load and speed Cooling/Exhaust system

temperatures/flowratestemperatures/flowrates Influence on powertrain control

Requirements for the Simulation: Engine model with thermal behavior Detailed cooling/ exhaust system model Powertrain model including friction

EM E-MachineTC Turbo ChargerHTC High Temperature Cooler



6

Model of control strategy softwareOC Oil Cooler

6


Development: Map-based ecoACC (HEV)


Development: Map based ecoACC (HEV)Function: Vehicle increases, decreases and

h ld d t lholds speed autonomously

Considered information:f

Navigation ACC

HMI

Desired vehicle behavior from HMI Track information from navigation Traffic information from radar

Map-basedecoACC

Requirements for the Simulation: Radar system model, traffic model

Powertrain Control

BrakeControl

Navigation system model Powertrain model Model of control strategy software• CO2-reduction

Cross domain function

Department | 06/09/2013 | © Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing, distribution, as well as in the event of applications for industrial property rights.


7

gy• Cross domain function

7


Contents

SimulationEnvironments








8


Optimization Thermal System Control Strategy (HEV)


Optimization Thermal System Control Strategy (HEV) load, speed

Plant ModelControl Software Model

Thermal systemLubrication

Exhaust System

actuator actuator sensor

temperatures, fuel consumptionmake

Tasks: Optimization of the thermal system (HW & SW) Testing of optimized SW version in a concept car

Rapid-Prototyping-HW



Testing of optimized SW version in a concept car

9


Combined HEV & Thermal Strategy Optimization


Combined HEV & Thermal Strategy Optimization

Control Software Model Plant Model

Driving cycle

(operation strategy thermal)actuatorsensor

Thermal systemLubrication

Exhaust System

Control Software Model Vehicle

(operation strategy hev)DLLImport

T k I ti t CO2 b fit f bi d th l d HEV t t ti i ti

temperatures, fuel consumption, SOC



Task: Investigate CO2 benefit of combined thermal and HEV strategy optimization

10


Simulation Environment for Map-based ecoACC


Simulation Environment for Map based ecoACC

Maneuver Road TrafficManeuver, Road,Traffic

ADAS RP

eHorizon actuator, sensorControl Software Model

Surrounding sensor info

Plant Model

Plant ModelPowertrain

GPS Position torque,speed

Tasks: Optimize & develop predictive powertrain control strategy fuel consumption, SOC



Tasks: Optimize & develop predictive powertrain control strategy

11


Contents

Parameterization &Validation



Used Simulation Environments





12


Parameterization


Parameterization

Perform special vehicle measurements

Use data from calibration departments

Use catalogue datageometric data

Vehicle.datEngineDyno

XLS

p

MotorDynoXLS

g

100

Script

0 50 100 150 200 250 3000

10

20

30

40

50

60

70

80

90

100

Vehicle Speed [km/h]

Acc

el. P

edal

Pos

. [%

]

shift12shift23shift34shift45shift56shift67shift78

Shiftstrategy



13


Validation: Fuel Consumption


Validation: Fuel Consumption

16

18Fuelconsumption NEDC

MeasurementSimulation

Rel. Error

MEAN 2.61 %

6

8

10

12

14

el c

onsu

mpt

ion

[ml/s

] SDEV 10.94 %

0 100 200 300 400 500 600 700 800 900 10000

2

4Fu

1000

1200

1400

1600

1800

onsu

mpt

ion

[ml]

Rel. Error

MEAN 1.71 %

SDEV 1.69 %

0

200

400

600

800

1000

Acc

umul

ated

fuel

co



0 100 200 300 400 500 600 700 800 900 10000

time [s]

14


Validation: Exhaust Temperature


Validation: Exhaust Temperature

900

1000Temperature downstream Catalyst

MeasurementSimulation

Rel. Error

MEAN - 1.89 %

S 10 31 %

700

800

900 SDEV 10.31 %

400

500

600

Tem

pera

ture

[°C

]

200

300

400

0 200 400 600 800 1000 1200 14000

100

time [s]



15


ContentsContents Scope & Motivation







16


Summary

Summary &Conclusion

Summary There are use case specific simulation environments for

Thermal Strategy Combined Strategy Predictive StrategyThermal Strategy Combined Strategy Predictive Strategy

GTSuite is applied for plant models

CarMaker is applied for radar sensor models & maneuver, traffic pp ,simulation

ASCET & Simulink are applied for control software models



ASCET & Simulink are applied for control software models

17


ConclusionConclusion GT Suite fulfills our requirements for a scalable simulation/modeling

environment for plant models

To cover all use cases several tools have to be integrated to one useful simulation application Requirement: Standardized interfaces for tool interoperability

Reuse of subsystem models coming from different tools is essentialRequirement: Standardized ex-/import features for model exchangeq / p g

Continue development in tool interoperability & model exchange features



Continue development in tool interoperability & model exchange features

18


Th k f tt ti !Thank you for your attention!



19

Co-Simulation for hybrid vehicle control software development · Co-Simulation For HEV Control...

Documents

Transcript of Co-Simulation for hybrid vehicle control software development · Co-Simulation For HEV Control...