E-Motor NVH - From Electric Current To Noise - tcm27-66019

E-Motor NVH – from electric
current to noise
Unrestricted © Siemens AG 2019 Realize innovation.
Electrification is here to stay
By 2040, we expect 57%

of all passenger vehicle
sales, and over 30% of
the global passenger
vehicle fleet, will be
electric
Source: Electric Vehicle Outlook 2019, BloombergNEF
Unrestricted © Siemens AG 2019

Page 2 2019-10-09 Siemens PLM Software
Electric Powertrain are not silent
Lower overall levels, but…

110.00 60.00
high-frequency, tonal content

makes e-motor noise
annoying:
gps:Speed (CH80)
dB(A)
km/h
Pa
• Higher motor orders due to
construction
• Very high frequency sounds
• Off-zero orders
• Related to PWM switching 10.00
50.00 Hz
Frequency
12000.00
-10.00
frequency

Addressing these E-Motor NVH challenges
– from electric current to noise
How to accurately predict e-motor

noise and identify system sensitivities
during the conceptual and detailed
design phase of vehicle development?

“V”- design cycle for automotive electric drives
Full Vehicle -
System Design
Subsystem EM design
Motor 3D Detailed Engineering

E-Powertrain Solution in the Subsystem Design Phase
Full Vehicle
System Simulation
Subsystem EM design

Impact on product life cycle and cost
Model-Driven Design to evaluate upfront systems designs
Ease of change Defects identification
availability
Cost to extract defects
FE model
availability
100%
Prototype
Model FE driven Prototype
driven design design driven design
50%
Design, Implement Delivery & Operation

Requirements Architecture Production
& Test Commissioning & Maintenance

availability
FE model
availability
100%
Prototype
Model FE driven
driven design design
Prototype
50% driven design


availability
FE model
availability
100%
Prototype
Model
driven design
FE driven
50% design
Prototype
driven design


Simcenter Amesim-based Noise App
Simcenter Amesim-based post-processing tool to

• estimate EM Forces from motor electrical model and control currents for given operating condition
• Predict radiated noise based on motor parameters and point source model
Offline 𝑀𝑎𝑔𝑛𝑒𝑡𝑖𝑐 𝐹𝑙𝑢𝑥 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 Geometrical

Electromagnetic equivalent
model System model parameters
Structural
model 𝑋(𝑓)
Look-
up Acoustic
𝐹(𝑡, 𝛼) model
𝑖𝑑𝑞0 𝑡 , 𝜃(𝑡)
𝐹𝑙𝑢𝑥 𝑙𝑖𝑛𝑘𝑎𝑔𝑒

Addressing e-Motor NVH at concept stage
Frontloading of qualitative NVH risk assessment
Providing NVH Teams an early tool to

assess influence of different electric
design variations on NVH

Subsystem Design Phase
Simcenter Amesim System Model
The e-machine model is integrated in a system-level

modelling environment. vehicle
Driver
Depending on application or maneuver, various
components can be connected within a multi-physical
cause/effect system of systems
Inverter
The e-motor produces torque based on EM maps and

Gearbox
control signals, but also the flux linkage and flux density
needed to calculate internal EM forces
MCU
E-motor
Battery

Electro-Magnetic Concept Model
Future vision
From actual EM model: From Analytical model::
EM model EM model Geometric / topology
parameters
.• Overall Geometry /
Topology
• Machine materials
Magnetostatic Analysis:
• Flux linkage maps for Id,
Iq and electrical angle System Model Build analytical
𝜃 𝑡 , 𝑖𝑑𝑞0 (𝑡)
force
𝐵 𝑡, 𝛼 = Λ 𝑡, 𝛼 × 𝑀𝑀𝐹(𝑡, 𝛼)
𝐹(𝑡, 𝛼)
𝐹(𝑡, 𝛼)

Vibro-Acoustic Model
The fluxes and forces from the EM model can be applied to simplified structural and acoustic models to
calculate the vibration and acoustic responses
Structural model Acoustic model
• Machine simplified
• Natural frequencies
geometry
• Modeshapes
• Machine materials
• Vibration response
• Fluxes / Forces
OUTPUT:
acoustic pressure

Design Modification Examples
Changing number of slots from 4 to 8 Changing stator geometry
48-4
48-8

Design Modification Examples
Inverter Modelling and Controls:

Simulink controller integrated with inverter/motor
Simcenter Amesim model
Purpose: evaluate control strategies to balance

switching noise & vehicle performance
Example: Random Carrier Frequency-PWM MCU / Inverter
(RCFM)

System-level design process for electric drives
Public reference of research project
Technical paper at SAE 2018 and publication in

SAE International Journal of Alternative Powertrains
Multi-Attribute, System-Level Design Process

for Automotive Powertrain Electric Drives: An
Integrated Approach
https://www.sae.org/publications/technical-
papers/content/08-07-02-0007/

Simcenter Amesim-based Noise App
Short demo:

Full Vehicle
System Simulation
Subsystem EM design

E-Powertrain Solutions in the Component Design Phase
From electromagnetics to radiated noise
Current Electro- Gear micro- eMotor, gearbox Gearbox Internals

Waveforms magnetic FEM geometry / TE and inverter
200 casing
Design Improvement /
Rpm = 1000
180 Rpm = 2000
Rpm = 3000
160 Rpm = 4000
Rpm = 5000
140 Rpm = 6000
Rpm = 7000
Optimization
120 Rpm = 8000
Current (A)
100
80
60
40
20
0
10 20 30 40 50 60 70 80 90 100
Rotor angle (deg)
EM Loads Transmission Structural and vibro-acoustic FE Model

Loads Weak Point Analysis
Correlation and
Updating
Casing Vibrations &
Radiated Noise
Prediction

Addressing e-Motor NVH in detailed design stage
End-to-end 3D simulation process
Providing NVH Teams an integrated

process to accurately predict e-motor noise
using state-of-the art tools and expertise

Focus on Electro-Magnetic field computation
Inverter Current Waveform

with MBD representation of
• Control Logic
• Power Electronics
• Motor Electrical resistance
Inverter Current waveforms
Electro-Magnetic Finite
Elements Model with well
established commercial FE
codes
• Fast analytical calculation
(MotorSolve)
• Accurate 2D/3D modeling
(MagNet) Transient magnetic forces Conversion Fourier components
Calculated on EM mesh Simcenter 3D
Infolytica MagNet

Focus on Transmission Loads
Typical process for NVH analysis

Transmission layout Transmission Error or Pre-processing of loads or
(stages, dimensions) Stiffness, parameters surface vibrations
Multi-body simulation Acoustics, NVH
Gear-centric tool
• Simulation of forces • Gear whine
• Analysis of gear pairs
and dynamics • Gear rattle
Positioning, dimensions…
More efficient process in Simcenter 3D

End-to-end integrated process for transmission
simulation from CAD to Loads to Noise
Transmission Builder → Motion → Motion-to-Acoustics
→ Acoustic Analysis
• Automatic creation of multi-body simulation models
• Accurate 3D simulation of gear forces
• Semi-automatic link of gear forces to vibro-acoustics
• Efficient and accurate acoustic simulations

Focus on Transmission Loads
Simcenter 3D Transmission Builder

A vertical application within Simcenter 3D: up to 5x faster Model creation process
Gear train specification Automated creation of Multibody

Simcenter 3D Transmission Builder
based on Industry standards simulation model

Transmission builder Validation
Transmission error validation for lightweight gears
Combined
friction,
microgeometry Low-frequency
and 280 Nm oscillation in
deformation both amplitude
effects on TE as global
240 Nm
behavior due to
body
180 Nm compliance
Validation cases ensure results (holes)
137 Nm
as accurate as non-linear Finite
Elements simulation 91 Nm
Microgeometry
matches well
Research (low loads)
45 Nm
partners:
Focus on Structural modeling
1. Rotor/Stator modeling - Laminated Structures Test EMA
Linear scalable CAE model with physical material

properties:
▪ Homogenization of orthotropic material

properties, for different area’s;
▪ Tuning of material properties based on test data
Simulation core components

Focus on Structural modeling
2. Windings Modeling – Discretization Test EMA with and without windings

Strategies
Linear CAE model with limited modal density
▪ Test based analysis of coils mass and stiffness

effects on e-Motor dynamics
▪ Dedicated coils modeling in CAE
▪ Tuning of material properties based on test data Simulation complete assembly

Focus on Vibro-Acoustics simulation
Acoustic Meshing with

dedicated tools
• Surface wrapping
• Convex Mesher
• Microphone Meshes
• …
Acoustics meshing
Vibro-Acoustics simulation
setup from Magnetic Loads
• Seamless workflow to re-
use magnetic forces from
Load recipes
electro-magnetic analyses
• Easy setup of combined
simulation for Dynamics
and Acoustics

Building a physical and reliable 3D model
Modal correlation – EMA testing Operational correlation– Bench/Vehicle testing
Component testing and updating; ▪ Experience – based knowledge of critical/sensitive

▪ Sub-assemblies testing and updating; parameters for operational response:
▪ Full system testing and updating ▪ Assembly tolerances and design robustness;
▪ Connections modeling and updating ▪ Non linearities;
▪ Screw connections; ▪ Stator and Rotor modeling (laminations vs.
▪ Welding connections; magnets and reinforcements);
▪ Press-fit connections

Rotor / stator modeling
Public references of research project
Technical paper at International Technical paper at International Technical paper at International

Conference on Noise & Conference on Structural Conference on Noise &
Vibration Engineering (ISMA) Engineering Dynamics (ICEDyn) Vibration Engineering (ISMA)
Experimental Study on the Experimental and Numerical Validation of homogenization

Impact of the Number of Validation of Laminated technique used for accurate
Laminas on the Dynamics Structure Dynamics from a predictions of laminated
Behavior of an Electric Switched Reluctance Machine structures’ mechanical
Machine Stator Stator behavior
https://link.springer.com/chapte https://lirias.kuleuven.be/174854 http://past.isma-

r/10.1007/978-3-319-30249- 7?limo=0 isaac.be/downloads/isma2016/
2_5 papers/isma2016_0379.pdf

Addressing e-Motor NVH in detailed design stage
Advanced Simulation topics
Providing advanced methods for NVH

optimization of controls as well as design
robustness

Vibration Synthesis
Full system optimization by use of model reduction technology

Vibration Synthesis
Full system optimization by use of model reduction technology
Advantage: Faster design of control strategies with full accuracy of the detailed EM and structural models
Potentially same accuracy as Can reach >10 times faster Insight in relevance of the
traditional 3D process calculations different spatial orders

Assembly tolerances and design robustness
Effect of rotor alignment on NVH performances
d Unbalanced Magnetic Pull (UMP) Effect of clearances & non-linearities
▪ Rotor misalignment results to unbalanced forces due to ▪ MBS model can be used to study:
non-uniform airgap distribution ▪ Assembly tolerances, pre-load settings
▪ Bearing clearances
▪ 3D coupled electromagnetic calculation requires huge
▪ Angular misalignments by using sliced approach
computational efforts
▪ Typical studies:
▪ Rattle phenomena;
→ Efficient process developed using a MBS model of the ▪ Evaluation of higher harmonics due to EM forces
rotor combined with table-based EM loads disturbances → Degradation of NVH performances

Summary
NVH assessment at concept stage 3D simulation in detailed design stage
• Brings NVH team to discussion table during • Integrated toolchain enabling design
electrical subsystem design phase improvement prediction on all aspects of e-
motor NVH: controls, EM, transmission,
• Multi-attribute balancing through system
structural
modelling approach
• High model quality and good correlation through
• Qualitative NVH risk assessment and quick
updating and parameter optimization process
evaluation of different design options
• Reduced-order model to enable fast controls
iterations
• Links to design robustness and manufacturing
tolerances

eMotor NVH – from electric current to noise
Questions?


E-Motor NVH - From Electric Current To Noise - tcm27-66019

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

E-Motor NVH - From Electric Current To Noise - tcm27-66019

Uploaded by

Copyright:

Available Formats

E-Motor NVH – from electric

By 2040, we expect 57%

Source: Electric Vehicle Outlook 2019, BloombergNEF

Unrestricted © Siemens AG 2019

Lower overall levels, but…

high-frequency, tonal content

Unrestricted © Siemens AG 2019

How to accurately predict e-motor

Unrestricted © Siemens AG 2019

Motor 3D Detailed Engineering

Unrestricted © Siemens AG 2019

Motor 3D Detailed Engineering

Unrestricted © Siemens AG 2019

Ease of change Defects identification

Design, Implement Delivery & Operation

Unrestricted © Siemens AG 2019

Ease of change Defects identification

Design, Implement Delivery & Operation

Unrestricted © Siemens AG 2019

Ease of change Defects identification

Design, Implement Delivery & Operation

Unrestricted © Siemens AG 2019

Simcenter Amesim-based post-processing tool to

Offline 𝑀𝑎𝑔𝑛𝑒𝑡𝑖𝑐 𝐹𝑙𝑢𝑥 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 Geometrical

Unrestricted © Siemens AG 2019

Providing NVH Teams an early tool to

Unrestricted © Siemens AG 2019

The e-machine model is integrated in a system-level

The e-motor produces torque based on EM maps and

Unrestricted © Siemens AG 2019

Unrestricted © Siemens AG 2019

Structural model Acoustic model

Unrestricted © Siemens AG 2019

Changing number of slots from 4 to 8 Changing stator geometry

Unrestricted © Siemens AG 2019

Inverter Modelling and Controls:

Purpose: evaluate control strategies to balance

Unrestricted © Siemens AG 2019

Technical paper at SAE 2018 and publication in

Multi-Attribute, System-Level Design Process

Unrestricted © Siemens AG 2019

Unrestricted © Siemens AG 2019

Motor 3D Detailed Engineering

Unrestricted © Siemens AG 2019

Current Electro- Gear micro- eMotor, gearbox Gearbox Internals

EM Loads Transmission Structural and vibro-acoustic FE Model

Unrestricted © Siemens AG 2019

Providing NVH Teams an integrated

Unrestricted © Siemens AG 2019

Inverter Current Waveform

Unrestricted © Siemens AG 2019

Typical process for NVH analysis

More efficient process in Simcenter 3D

Unrestricted © Siemens AG 2019

Simcenter 3D Transmission Builder

Gear train specification Automated creation of Multibody

Unrestricted © Siemens AG 2019

1. Rotor/Stator modeling - Laminated Structures Test EMA

Linear scalable CAE model with physical material

▪ Homogenization of orthotropic material

Unrestricted © Siemens AG 2019

2. Windings Modeling – Discretization Test EMA with and without windings

Linear CAE model with limited modal density