Exploring thousands of configurations: Find the best design out of infinite variants with LMS System Synthesis

Config_1_1 Config_1_2 Config_1_3
Exploring thousands
of configurations
Find the best design out of infinite variants
with LMS System Synthesis
Realize innovation.Unrestricted © Siemens AG 2017

Unrestricted © Siemens AG 2017
Page 2 Siemens PLM Software
Agenda
• Overview of LMS System Synthesis
• Multi-attribute balancing and
variant analysis
• Systems-driven product
development (SDPD)
• Applications
• Going further
• Definitions
Best range = 206 km
Best efficiency
= 0,1125kW/ km
Best power &
Performance ratings

Engineering decisions intrinsically involve systems
Design – Configuration – Physical behaviors – Electronics, controls and SW
Complicated
Segregated systems / divisible
Complex
Integrated systems / non-divisible
• Model-based
systems engineering
• Integrated processes
• Integrated verification
• Document-based
systems engineering
• Isolated departments
• Distributed verification
System of systems
Interdependent – extra-vehicle
• Enterprise systems engineering
• Systematic integration
• Continuous verification
• Multi-attribute performance
Multi-attribute balancing
Vehicle integration
Driving
Dynamics
NVH and
Acoustics
Structure
& Durability
Thermal
and Energy
Safety
Chassis and
Suspension
Body Powertrain E&E
Full
Vehicle

Model-based systems engineering challenges
Dealing with many variants
2160 configurations to be
evaluated
These models could be represented
through
1 architecture and 48 component models
Vehicle
(2 platforms)
Battery
(3 types)
…
…
Tires
AC motor
….
(2x2 each)
Transmission
(2 types AT/MT)
Engine
(5 types)
Model
accuracy
• Concept
models
• Measured
data
• Detailed
models
Load cases
• EU
• Asia
• US

ConfigurationSimulation Architecture
Deployment of
system engineering
Teamcenter – LMS Imagine.Lab
Product suite and positioning in systems engineering
Product Life Management - Teamcenter
Stand Alone
or PLM Plugin
Functional
architecture
LMS Amesim
Other CAE disciplines
Engine
Specialist
Chassis Specialist Controls
Specialist
Transmission
Specialist
LMS System Synthesis
Requirements
Functions
Logical
Physical
PLM platform

Concept of simulation factory for the sake of the simulation engineer
• Few experts are creating
models
• Waste of time and ressource
in « repetitive task »
• Common background for
company
• Common infrastructure for a
program
• Design rules are enforced
from expert to engineers
1-10
~50
~100
>100
System
engineer
Simulation
architect
Model expert
Calibrator Model
user

Simcenter™ Portfolio for Predictive Engineering Analytics
LMS Imagine.Lab

Agenda
variant analysis
development (SDPD)
• Applications
• Going further
• Definitions
Best range = 206 km
Best efficiency
= 0,1125kW/ km
Best power &
Performance ratings

Architecture-driven simulation
VCU
Driver
Vehicle
Gearbox
Electric
motor
Battery
• Tool neutral architecture consisting
of templates defining the interface
contract between subsystem
behavioral models
• Easy integration of subsystem
models from different departments
• Execution of heterogenous
simulation architectures
LMS Amesim LMS Amesim
LMS Amesim
LMS Amesim

VCU
Driver
Vehicle
Gearbox
Electric
motor
Battery
behavioral models
• Enable scenario and variant
management

behavioral models
• Enable scenario and variant
management
• Perform multi-attribute balancing
within one project
Performance
simulation architecture
Comfort
simulation architecture
Base architecture
ComfortPerformance
Attributes

Agenda
variant analysis
development (SDPD)
• Applications
• Going further
• Definitions
Best range = 206 km
Best efficiency
= 0,1125kW/ km
Best power &
Performance ratings

Systems-driven product
development (SDPD)
Key challenges driving our strategy
Optimize competing
targets
Manage product data
complexity
Enable systematic reuse
Balance performance and
quality attributes
Consistency
Coordination
 Existing tools
 Existing processes
 People
Integrate and coordinate
engineering disciplines

Model-based systems engineering (MBSE) workflow
Unstructured process
2
System
simulation
I need a model
Advanced
engineering
departments
Local hard discs
Local
servers
Stored for department need only
unclassified, no meta-data,
no corporate access control
1
This model
should be ok.
2
This test data
should be ok.
Unclassified, no meta-data,
no corporate access control
Risk analysis - Model generation
• Confusion prone
• No overview
• No control
1
Risk analysis - Model exchange
• No standard workflow
• No security
• No traceability
2
Risk analysis - System simulation
• Manual work
• Error prone definition
• Error prone traceability
3
Simulation results generation
3

MLM
Model lifecycle management
Many challenges remain for system simulation engineers
Advanced
engineering
departments
Pool of classified
models stored on one
repository
Model management
& meta-data client
1
Search
based on
meta data
32
Import or create
Dedicated models
Execute
System simulation
with available models
Manage system
configurations
for traceability
System
simulation
Which
parameter?
Which models
to use?
Where to store
optimized
parameters
How to track
results?
How to do trade-
off study?
What are the
targets?

MLM
Model Lifecycle Management
Closed-loop
performance
engineering and
system design
Closed loop performance engineering and system design
Advanced
engineering
departments
Pool of classified
models stored on one
repository
Model management
& Meta-data client
Search
based on
meta data
System
simulation
System
design
Take
Design decisions
based on performance
assessment
Target
Input
parameter
ModelResults
Optimized
parameter
Target
Input
parameter
Model
Results
Optimized
parameter

Product Lifecycle Management
Requirements Logical PhysicalFunctional
Architecture Mapping Deploy simulationAnalysis request
LMS Amesim
1D CAE
Data management: LMS Sysdm, Teamcenter
LMS System Synthesis 15 overview
LMS platform in the Product Lifecycle Management (PLM) environment
Teamcenter

Program director, interested in:
• Schedule
• Open engineering changes and issues
• Cost of program
Vehicle system architect, interested in:
• Vehicle level architecture
• Functional system integration
• Open engineering changes and
issues
Product configuration specialist, interested in:
• Features
• Variability
• Marketing
Requirements engineer, interested in:
• Documentation of requirements
• Validation of requirements
Domain architect, interested in:
• Architecture of a system, like HVAC
• Requirement
• System functions
• System behaviors
Domain engineer, interested in:
• Control modeling
• Source code management
• Calibration and configuration
The product line engineering factory

Simulation architecture
• Abstract description of the system
• Tool neutral
• Not simulatable / no behavior information
• System structure description
• Model template
• Connection between model template
A simulation architecture is defined
• From a logical architecture
• By refactoring the logical architecture structuring
• By adding simulation specific information
• According to a simulation purpose
• Fuel efficiency
• …
Architecture

User A
Individual users create collections
associating with other « Core » collections
Core set of collections
User B
Research program 2025
Hybrid program 2017Engine
Gearbox
Vehicle
Electric motor
Battery
Central repository
Re-usability of core models / project

Agenda
variant analysis
development (SDPD)
• Applications: for automotive
• Going further
• Definitions

Variants evaluation
NEDC WOT Loaded
slope
Battery #1 Battery #2 Battery #3
EM #1 EM #2
Attributes
Range Efficiency Performance
Continuous PowerPeak Power
Scenario’s
Variants
Architecture
0,12 kWh/km150 km 140 km/h
50 s 80 km/hVCU
Driver
Vehicle
Gearbox
Electric
motor
Battery
Case of the electric vehicle (EV)
Overview of the whole picture

Architecture and template definition
Architecture &
template definition
Model
instrumentation
Model
assembly
Study & run
execution
Variant evaluation
Architecture
• Define subsystems
• Define the connections /
relations between subsystems
Templates
• Defined for each subsystem
• Interface contract (i.e.
Input/output definition)
• Parameters and variables
Architecture
Templates
Battery
i
SOC
V

Model instrumentation
Architecture &
template definition
Model
instrumentation
Model
assembly
Study & run
execution
Variant evaluation
Behavioral model connections
• Complex interactions involve
multiple “wires” between
subsystems
• No interface contract defined
• Limited re-use conditions
Instrumented model connections
• Interface contract definition
• Subsystems are connected using
a single connector
• Increased re-use of subsystems

Model assembly
Architecture &
template definition
Model
instrumentation
Model
assembly
Study & run
execution
Variant evaluation
Model assembly
• Select for each template an
instrumented model
• “Plug & play” connection of
models thanks to interface
contract
Scenario
EM Gearbox
Controls Vehicle
Battery
Model assembly

Study & run execution
Architecture &
template definition
Model
instrumentation
Model
assembly
Study & run
execution
Variant evaluation
Run
• Defines a single simulation
scenario (example: NEDC cycle)
Study
• Groups several runs
Scenario’s Attributes
NEDC
WOT slopeLoaded
Range Efficiency
0,12 kWh/km150 km
Performance
Continuous PowerPeak Power
140 km/h
50 s 80 km/h

Study: running multiple scenario’s to score on different attributes
NEDC WOT Loaded slope

Variants
Architecture &
template definition
Model
instrumentation
Model
assembly
Study & run
execution
Variant evaluation
Run
• Defines a single simulation
scenario (NEDC Cycle )
Study
• Defined for each variant
• Groups all scenario’s for this
variant
Battery #1 Battery #2 Battery #3
EM #1 EM #2
Variants

Variants
• Max Torque = 208 Nm
• Efficient
• Power Loss Map:
EM #1
• Max Torque = 215 Nm
• Less efficient
• Power Loss Map:
EM #2
• Energy = 16.5 kWh
• Voltage = 386 V
• Mass = 118 kg
Battery #1
• Energy = 21.8 kWh
• Voltage = 336 V
• Mass = 156 kg
Battery #2
• Energy = 24 kWh
• Voltage = 360 V
• Mass = 171 kg
Battery #3
EM behavioral models impemented as
FMU’s for co-simulation (FMI 2.0)
Battery behavioral models impemented as seperate
LMS Amesim supercomponents

Configurations and evaluation
EM #1
EM #2
Bat #1 Bat #2 Bat #3
Torque
Efficiency Energy
Best range = 206 km
Best efficiency
= 0,1125kW/ km
Best power &
Performance ratings
Best balanced

Agenda
variant analysis
development (SDPD)
• Applications: for aerospace
• Going further
• Definitions

Sub-systems models and
tools
Electric
aircraft
Aircraft
engine
Environm.
control
system
Fuel
systems
Engine
equipment
Flight
control
Landing
gear
LMS Imagine.Lab platform
LMS Amesim
System model – Synthesis and analysisScenarios Performance

LMS Imagine.Lab platform
System model – Synthesis and analysisScenarios Performance
Manage and execute
multiple scenario’s.
Ensure integration of behavioral models
created in different tools
by different departments.
Manage and
evaluate multiple
variants.
Enable multi-
attribute
balancing.
Template
Scenario 1
Scenario 2
Scenario 3
…
Attribute 1
Attribute 2
Attribute 3
…
Variant 1
Variant 2
Variant 3
…

Application case: aileron actuation system
• Actuation system architecture
consisting of templates defining the
Interface contract between
subsystem behavioral models
ActuationController Body
Propulsion
Controls
Motor
Aileron
EHA
LMS AmesimLMS Amesim

• Execute different scenario’s and
ambient conditions
• Analyze different variants: fluids
properties, electric motor
technologies, …
• Analysis: pressure in the jack for
different fluids
Scenario 3
Motor
Aileron
EHA
Scenario 2
Scenario 1
EHA
EHA
Aileron
Aileron
Motor
Motor
Different
scenario’s
Different
fluids
ActuationController Body
Propulsion
Jack Pressure [bar]

• Perform technology trade-off and
multi-attribute balancing within one
project
• Compare different architectures EHA
vs. EMA
• Analysis: actuation force evaluation
for EHA and EMA architecures
Base architecture
EMA architecture EHA architecture
EHA
EMA
Technologies
Actuation force [bar]

Agenda
variant analysis
development (SDPD)
• Applications
• Going further
• Definitions
Best range = 206 km
Best efficiency
= 0,1125kW/ km
Best power &
Performance ratings

LMS System Synthesis documentation in LMS Amehelp
Concepts and
examples of use
LMS System Synthesis 15 online documentation comes with
its user guide and tutorials to start with:
• Architectures,  Configurations
• Variants,  Analysis requests (Teamcenter)
…

LMS Imagine.Lab landscape
Collaborative platform supporting model authoring, configuration and integration
Authoring Behavioral
Models
LMS Amesim
IT Admin
E
C
U
Vehicle Energy
Management
Virtual Integrated
Aircraft
Configuration,
Integration
Siemens model & data
management
MLM (Model Lifecycle Management) « Service Data Bus »

System architecture & modeling
Cross domain validation
Engineering domains realization of system logical architecture
Closed-loop with cross-domain asset management
• Parameters
• Targets
• Attributes
• Models
• Test data
• Simulation results
Behavior modeling/simulation
Integrated detailed design
Domain architecture
Sensor
SWC
Software component architecture
[component architecture model]
Control
SWC
Actuator
SWC
Control
SWC
Control
SWC
Actuator
SWC
ECU
ECU
Gateway
Bus
Bus
Sensor
Actuator-L
Actuator-R
Electronic/electrical architecture
[network architecture model]
deploy
Behavior [behavior models]
Mechanical modular architecture
[3D DMU model]
Multi-domain simulation architecture
[behavior simulation architecture model]
DynamicsEnergy
conversion
Plant
Control
Mechanism
realize
Control 1D- Plant 3D- Plant CAEAnalysis Calibration Diagnostic FMEAElectronics
…
configure
specify specify specify
behaviorbehaviorbehavior
use use
System architectureRFL
System model
realize

Agenda
variant analysis
• System-driven product
development (SDPD)
• Applications
• Going further
• Definitions
Best range = 206 km
Best efficiency
= 0,1125kW/ km
Best power &
Performance ratings

Definitions
• Architecture: An architecture describes the generic structure for a
simulation model. It is a blueprint for the « to be assembled »
simulation model.
• Attribute: A property or characteristic of one or more entities.
• Base architecture: High level representation of the future model to
set. It defines the topology of the system model. The base
architecture is a reusable structure that allows to bridge the
simulation world to the system engineering one. It describes the
overall simulation structure issued from the system model.
• Behavioral model: A component or subsystem object can contain
multiple behavioral models, describing the component or
subsystem at different levels of detail. It is possible to capture this
set of behavioral models in a single reconfigurable model. This
requires, however, that all behavioral models have the same
interface, a condition.
Base architecture
Simulation architecture

Definitions
• Configuration: A configuration associates instrumented models to
the system model architecture.
• Exposure: The exposures are non directional numerical properties,
they act as a first level of contract definition for the template.
• Instrumentation: An instrumented model is a reusable artefact that
links a simulation template to a behavioral model by identifying the
relations between parameters, variables and ports.
• Interface contract: All the ports combined are the interface of the
model. The interface contract defines how the component can
interact with the other components in the system, but does not
contain any information about the internal behavior of the
component.
• Model assembly: A model assembly is the configuration of a
simulation architecture with the set of appropriate instrumented
model. The model assembly describes the numerical process to be
realized in order to provide the results to the simulation engineer.
Automotive template library
Base template
Base port
Exposures

Definitions
• Simulation architecture: Defines the interface of the system model.
The simulation architecture describes how results are to be
created, what are the variables to be exchanged between the
elementary pieces of the assembly.
• Simulation model: A simulation model is the final artefact that
allows a computer to produce the expected set of numerical values
that are considered as a « result » by the simulation engineer.
• Template: The templates are the elementary pieces used to
describe the architectures. Stored in libraries they are reusable
across different architectures and across projects in a few
properties.
• Variant: A variant is a specialization of a model and inherits all of
its properties, unless they are defined locally. This allows a very
compact modeling of models that vary only.
Simulation
template
Behavioral
model
Instrumented model

Realize innovation.
Nico VANSINA
Loig ALLAIN
Stéphane NEYRAT
LMS Amesim
Siemens Industry Software S.A.S.
Digital Factory Division
Product Lifecycle Management
Simulation & Test Solutions
DF PL STS CAE 1D

Exploring thousands of configurations: Find the best design out of infinite variants with LMS System Synthesis

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (6)

Similar to Exploring thousands of configurations: Find the best design out of infinite variants with LMS System Synthesis

Similar to Exploring thousands of configurations: Find the best design out of infinite variants with LMS System Synthesis (20)

Recently uploaded

Recently uploaded (20)

Exploring thousands of configurations: Find the best design out of infinite variants with LMS System Synthesis