dr. mihai nica dvp&r...dr. mihai nica florian klück m.sc. avl list gmbh (headquarters)...
TRANSCRIPT
Dr. Mihai Nica
Florian Klück M.Sc.
AVL List GmbH (Headquarters)
Confidential
DVP&R
DVP&R Methodology @ AVL
Validation of Electrified Powertrain Systems
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 2Confidential
DVP&R
The AVL-PTE Design Verification Plan & Reporting (DVP&R)unifies in one Workproduct (Master-Document) the Validationand Verification (V&V) test procedures that are to beperformed in the context of development projects ofmechatronic systems or its subsystems.
Additionally it enables progress monitoring anddocumentation of the execution of the V&V activities withinthe project development.
In this context the DVP&R combines the validation andverification principles of
mechanics,
electric & electronics,
and software development (including calibration)
to generate a simpler, more economical and reliable validationand verification process.
DVP&R: Unified test program for all skill areas
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 3Confidential
• DVP challenge and AVL solution
• AVL DVP methodology
• TARGETS and LOAD PROFILING
• Test Program
• Evaluation & Optimization
• Summary & Conclusion
AGENDA
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 4Confidential
DVP CHALLENGE & AVL SOLUTION
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 5Confidential
“Are the responsibilities
aligned?”
Traditional development process & test programs
New Worksplitbetween
departments
Modified responsibilities
New WorksplitSupplier &
OEM
Economic boundaries
Timeline
Budget
Capacity
DVP CHALLENGES
“Are we testing the right things?”“Do we have the right targets?”
“Are we testing at the right
time?”
New markets/ applications
Workflow Responsibilities Test targets Testing approaches
Specification & Requirements & Milestones
New/ modified systems
Verification tasks?
Failure modes?
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 6Confidential
Evaluation
FP - Analysis
Load Analysis
Mission Profile
Results, failures, measurement, data, new procedures
Test program
functional
development
Functional
development
and
reliability
&
durability
testing
• Basic DVP
• FMEA
• FP -Database
• Technical
specifications
• Others
Function
Reliability
&
Durability
Functional
issues
Design Verification & Validation Plan (DVP)Design Verification & Validation Plan (DVP)
Durability
&
reliability
test program
Reliability
and
durability
issues
LOAD MATRIX TM
Testing
requirementsTesting
Optimization
FP- Analysis
Load Analysis
Targets
Results, failures, measurement, data, new procedures
Test program
functional
development
Functional
development
and
reliability
&
durability
testing
• Basic DVP
• FMEA
• FP -Database
• Technical
specifications
• Others
Function
Reliability
&
Durability
Functional
issues
Design Verification and Validation Plan & Reporting (DVP&R)
Initial DVP durability
test program
Reliability
and
durability
issues
LOAD MATRIX
Testing
requirementsTesting
AVL DVP&R PROCESS & LOAD MATRIX INTERACTION
Verification
Validation
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 7Confidential
AVL DVP METHODOLOGY
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 8Confidential
DVP&R LIFE CYCLE
② DVP&R Planning
③ DVP&R Monitoring
① DVP&R Target
Definition
System Analysis
Application & Targets
Test Program
Evaluation andOptimization
System Analysis
Application & Targets
Test Program
Evaluation andOptimization
System Analysis
Application & Targets
Test Program
Evaluation andOptimization
System Analysis
Application & Targets
Test Program
Evaluation andOptimization
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 9Confidential
DVP&R PROCESS FLOW
For each project, the DVP&R Work product (i.e. Deliverable) will be planned and monitor on a level - based approach
(similar to the requirement engineering process)
……
EHW
…
SWSHW
PartsS+A …
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 10Confidential
IDENTIFY VERIFICATION TASKS FROM VEHICLE LEVEL DOWN TO COMPONENT LEVEL
System analysis
Generate a common understanding of the covered technical elements and their interfaces
DIFF
Electric Powertrain Example Topology Motor, Inverter
(+ HV Harness)
Shafts• Cardan shaft• Drive shaft• Rigid axle
• Transmission
Reduction Gear Unit
Electric Powertrain System:
- Motor
- Inverter
- HV Harness
(Inverter -> Motor)
- Transmissions
- Cardan and Drive Shafts
- Axle/Differential Gear Unit
- (Rigid) Axle
Axle/Differential Gear Unit
(Rigid) Axle
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 11Confidential
VISUALIZE VERIFICATION TASKS & FAILURE MODESEXAMPLE
Differential Gear Unit
Cardan & drive shafts
(Rigid) Axle
Transmission
HV-Harness
Motor
Inverter
Component failure mode combinations Verification tasks
Tasks to be executed
Tasks with unknown impact
Failure modes with unknown impact
Failure modes relevant
for verification
Failure modes relevant
for verification & validation
System analysis
Generate an overview what needs to be covered by the DVPDefine countermeasures to clarify unknown impacts
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 12Confidential
AVL DVP METHODOLOGY
TARGETS and LOAD PROFILING
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 13Confidential
LOAD PROFILING – USAGE SPACE ANALYSIS
Load
Dynam
ics
ExtraUrban
Highway
Offroad
Urban
Load
Dynam
ics
Field driving cycles
Analyze customer driving
and vehicle usage
patterns
Develop representative
customer vehicle usage
cycles
Profiles which include
velocity and gradient
Analyze target market
Analysis regarding:
Geography
Driving behavior
Road characteristics
Market segments
Population
Time
Velo
city
Example: Reference Highway
Analyze ambient
conditions
Analysis regarding:
Different regions
Different seasons /
months
Representative
distributions
Example: Ambient temperature distributions in
different areas
Example: Population of China
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 14Confidential
LOAD PROFILING – USAGE SPACE ANALYSIS
Lifetime distribution
matrix
No pre-existing data
Market researches
Varying boundaries
(vehicle loading, vehicle
gradients, ambient
conditions, vehicle modes)
Reduction step
Set up weighting tables
Calculate lifetime mileage
proportion
Consider variations which
fulfill target
Example: Lifetime distribution matrix
Generate larger data
base
Design of experiments
Variations included
Full variation of cycles &
parameters would lead to a
high number of cycles Example: Reduction step
TOTAL Urban Rural Highway Offroad
100.0% 20.0% 60.0% 18.0% 2.0%
10.0% 50.0% 11.0% 1.0%
10.0% 10.0% 7.0% 1.0%
15.0% 55.0% 15.0% 0.0%
5.0% 5.0% 3.0% 2.0%
4.0% 5.0% 4.0% 0.0%
14.0% 50.0% 12.0% 2.0%
2.0% 5.0% 2.0% 0.0%
100.0%
100.0%
100.00%
Vehicle loading: full
Ambient Temperatur (very low)
Ambient Temperatur (standard)
Ambient Temperatur (very hot)
%(lifetime km / lifetime hours)
Overall
Vehicle Gradient: flat
Vehicle Gradient: hilly
Vehicle loading: empty
TOTAL Auto Power Pure 4WD Save
100% 60% 5% 10% 20% 5%
(% of time)
Vehicle Modes
Flat
Hilly
Example:
profile slope weight
urban flat 15.0%
urban hilly 5.0%
rural flat 55.0%
rural hilly 5.0%
highway flat 15.0%
highway hilly 3.0%
offroad flat 0.0%
offroad hilly 2.0%ambient weight
very low 13.0%
standard 78.0%
very hot 9.0%
profile slope loading ambient mode weight
urban flat empty very low auto 0.585%
rural flat empty very low auto 3.575%
highway flat empty very low auto 0.715%
offroad flat empty very low auto 0.000%
urban hilly empty very low auto 0.195%
rural hilly empty very low auto 0.325%
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 15Confidential
DEFINITION OF COMPONENT LOAD PROFILES DERIVED FROM REAL WORLD VEHICLE DATA
Results:
• Reference cycles based on actual end customer usage
• Market related analysis report
Results
• Time based component load data• Parameter variation covering different
ambient conditions or control strategies
Results
• Lifetime load profiles • Lifetime load collectives for
components
Simulation model
Failure parameter
sheet
Damage modeling & statistical analysis
Usage space analysis
Reference cycle(s)
Targets & responsibilities
Utilize customer fleet data efficiently to generate load profiles for new systems and components
Fleet data
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 16Confidential
AVL DVP METHODOLOGY
TEST PROGRAM
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 17Confidential
DVP&R: GENERATION OF V&V PLANS
Software
…Design
Desig
n
Cha
nge
s
System
Qu
alit
y G
ate
s
Mile
sto
ne
s,
Te
sts
RCL
DVP&R
FP-Sheet
FMEA
Boundary
Diagrams
Requirements
Interface Analysis
Quality History
Customer Wants
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 18Confidential
DVP&RMAIN TEST CATEGORIES
10 Functional System Development and Integration
20 Geometrical Integration
• Performance and range• Core Functions (e.g. external Charging)• Emission and Consumption System• On-Board Diagnostics
• Geometrical Targets (Digital Mock-Up)• Weight monitoring• Center of gravity and inertia• Assembly (e.g. accessibility)• Serviceability
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 19Confidential
DVP&RMAIN TEST CATEGORIES
30 Mechanical System Development and Integration
• Durability• Mechanical Functional Testing (e.g. Efficiency)• NVH Functional Testing (e.g. NVH Analysis of
e-Machine)• Robustness and Misuse• Vibration, Stiffness, Strength• Shock, Crash, Crush
• Condensation and pressure handling• Cooling system performance• Tightness and vacuum test• Filling and Degassing of cooling circuit• Wind Tunnel Testing• Street testing (e.g. uphill towing)
40 Thermal System Development and Integration
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 20Confidential
DVP&RMAIN TEST CATEGORIES
50 Electrical System Development and Integration
60 System Safety
• HV Safety (ISO 26262)• Functional Safety• Vehicle Safety (active and
functional passive safety)• Safety of components
• HV Architecture• Energy Management• Load Balance• EMC (e.g. Immunity, Emission, ESD)
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 21Confidential
ADAS test and validation tools
ADAS V&V
ADAS validation (MIL cloud, MIL office, HIL,VIL, proving ground, public street)
Develop ADAS functions, control units and vehicle integration
ADAS Center
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 22Confidential
AVL DVP METHODOLOGY
EVALUATION & OPTIMIZATION
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 23Confidential
VISUALIZE STATUS AND RISKS OF ACTUAL DVP
Differential Gear Unit
Cardan & drive shafts
(Rigid) Axle
Transmission
HV-Harness
Motor
Inverter
Required tasks from system analysis Correlation with planned test program
Tasks to be executed
Tasks with unknown impact
Test programConnect verification tasks and failure mode with the planned program
Visualize risks and guide team to close gaps
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 24Confidential
DEFINITION AND VISUALIZATION OF DVP IMPROVEMENTS
Applied improvements
Targets & timeline Specify targets Improve timing of testingResponsibilities Align responsibilities
“Achievebest possibletask coverage”
Test program Modify procedures Define new proceduresRisk based task coverage Decide risk based on tasks in development
team
Status DVP-Result Main request
First analysis results
“Clarify responsibilities
Andtargets”
1st update of analysis
Final update of analysis
“Visualize effects of changes”
Management and project team Agree on changes Highlight & agree on remaining risks
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 25Confidential
SUMMARY & CONCLUSION
Dr. Nica Mihai, Florian Klück M.Sc. | | 05 Juli 2017 | 26Confidential
AVL DVPBENEFITS
AVL DVP
Increased technical and economical efficiency of verification and validation test program
Minimized risk of verification & validation gaps
Consideration of individual program targets, boundaries and customer processes. Evaluation and
optimization
Collection of all required testing activities in one document
Correlation of verification tasks and failure modes with planned testing program
Clear definition of project specific verification & validation targets
Alignment of responsibilities for complete test program Highlight needed missing information
Systemanalysis
Targetsand
Responsibilities
Test Program
Clear visibility of the verification & validation risks for all relevant components
Structured approach to make progress & improvements visible
Identification of verification tasks from vehicle level down to component level
Identification of components and failure modes to be covered by testing in one document