virtual vehicle application: battery cooling network study · 3 key takeaways battery cooling...
TRANSCRIPT
![Page 1: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/1.jpg)
1© 2020 The MathWorks, Inc.
Virtual Vehicle Application:
Battery Cooling Network Study
![Page 2: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/2.jpg)
2© 2020 The MathWorks, Inc.
Virtual Vehicle Application:
Battery Cooling Network Study
![Page 3: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/3.jpg)
3
Key Takeaways
▪ Battery cooling network design
requires component level analysis
and tests within a full-vehicle simulation
▪ Integrating fluid, thermal, electrical,
and mechanical domains is key
to assessing system-level performance
▪ Rapid simulations covering a wide range
of drive cycles and ambient conditions
are needed to evaluate design criteria
Lumped Parameter Network
CFD and FEA
Computation
Time
Model Fidelity
Simscape
Model
Powertrain Blockset Model
Spreadsheet
![Page 4: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/4.jpg)
4
Agenda
▪ Importance of Battery Cooling
▪ Exploring Battery Cooling Network Designs
▪ Integration in Vehicle Model
▪ Evaluation of Design in Full Vehicle Tests
![Page 5: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/5.jpg)
5
Why Explore Battery Cooling?
▪ Electrification is a cross-industry market driver
– Power, heating, transportation
– Shift to electric and hybrid powertrains
▪ Key to success: efficiency and safety
Source: Technische Universität München
Voltage vs. Charge
Charge (Ah)
Safety Concerns
Thermal
Runaway
Leads to
Explosion
EV Sales and Market ShareData Source: International Energy Agency (2018)
2014 2016 2018Battery State of Health
Time (Days)
Source: Science (Feb 05 2016)
Too Hot:
Efficiency
Reduced
By 50%
Too Cold:
Output
Reduced
By 80%
2013 2015 2017
![Page 6: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/6.jpg)
6
Agenda
▪ Importance of Battery Cooling
▪ Exploring Battery Cooling Network Designs
▪ Integration in Vehicle Model
▪ Evaluation of Design in Full Vehicle Tests
![Page 7: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/7.jpg)
7
Design Challenge: Battery Cooling Network
▪ Requirements
– Cell temperature range: 20-40 °C
– Cell temperature max delta: 8 °C
▪ Evaluation
– Hot and cold environments
– Driving conditions (FTP75, US06, WOT, etc.)
– Charge cycle
▪ Two options considered
– One-pass
– Two-pass
One-Pass
Two-Pass
< 8°C
20°C
40°C
![Page 8: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/8.jpg)
8
Design Process for Battery Cooling Network
1. Explore designs
![Page 9: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/9.jpg)
9
Design Process for Battery Cooling Network
2. Integrate in vehicle model
1. Explore designs
![Page 10: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/10.jpg)
10
Vehicle
Vehicle-level
Controllers
DriverManeuver/
Drive Cycle
Environment
Design Process for Battery Cooling Network
2. Integrate in vehicle model
3. Perform full vehicle tests
1. Explore designs
![Page 11: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/11.jpg)
11
▪ CFD and FEA
– Accurate, but computation intensive
▪ Spreadsheet
– Accessible, but limited scalability
– Limited options for integrating other models
▪ Lumped parameter physical networks
– Less accurate than CFD, but scalable
– Appropriate for system-level analysis
– Integrates well with other domains
including control algorithms
Com
puta
tional
Tim
e
Model Complexity & Detail
Computational Time vs. Model Complexity
Lumped
Parameter
Network
Spreadsheet
CFD and
FEA
Design Challenge: Battery Cooling NetworkModeling and Simulation Options
![Page 12: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/12.jpg)
12
Simscape: Build Accurate Models Quickly
▪ Simply connect the
components you need
▪ The more complex the
system, the more value
you get from Simscape
▪ Resulting model is
intuitive, easy to modify,
and easy for others
to understand
FSpring = kSpring *(xMass)
FDamper = bDamper*(dxMass
dt)
d2xMassdt2
=−FSpring − FDamper
mMass
Input/Output Block Diagram Simscape
![Page 13: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/13.jpg)
13
▪ Simulink is best known for
signal-based modeling
– Causal, or input/output
▪ Simscape enables
bidirectional flow of energy
between components
▪ System level equations:
– Formulated automatically
– Solved simultaneously
– Cover multiple domains
Physical Modeling
Within Simulink
R1
C1v1
i+
_
R1
C1
i2
+
_i3
i1 R2
C2v2
Simulink: Input/Output
Simscape: Physical Networks
![Page 14: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/14.jpg)
14
Battery Model
▪ Modeled using Simscape
– 60kWh total capacity (4 sections)
– Equivalent circuit captures transient dynamics
– Lookup tables: nonlinear and thermal effects
– Battery aging can be included
Resistors, capacitor, and voltage source
depend upon SOC, DOC, and temperature
![Page 15: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/15.jpg)
15
Battery Pack
▪ Create test to compare the
cooling network designs
▪ Lumped thermal model
– Divided into four sections along flow path
▪ Heat transferred to different
portions of the cooling channel
![Page 16: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/16.jpg)
16
Battery Cooling Network
▪ Physical connections in the Simscape model match architecture of design
One-Pass Two-Pass
![Page 17: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/17.jpg)
17
Battery Cooling Network
▪ Simplify testing using
Variant Subsystems
– Swap in different cooling designs
– Interactive or automated
using MATLAB commands
▪ Same model, settings,
and test set up
– Input vectors
– Results analysis
Variant
Two-Pass
One-Pass
![Page 18: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/18.jpg)
18
Cooling Network Test
▪ Fast charge (cooling critical)
1. From 2% to 99% in 1 hour
2. Range of coolant flow rates
1
2
![Page 19: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/19.jpg)
19
Cooling Network Test
▪ Fast charge (cooling critical)
▪ Performance criteria
a. Maximum temperature
b. Temperature gradients
c. Pump power consumption
a
< 8°C
< 40°C20°C<
b
c
1
2
![Page 20: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/20.jpg)
20
Component Level Analysis
▪ Criteria 1: Temperature Range
– For same flow rate, Two-Pass
has lower maximum temperature
– Acceptable range for either design
< 40°C20°C<
![Page 21: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/21.jpg)
21
Component Level Analysis
▪ Criteria 1: Temperature Range
– For same flow rate, Two-Pass
has lower maximum temperature
– Acceptable range for either design
▪ Criteria 2: Temperature Gradient
– Both designs acceptable
– Two-pass has very low temperature
difference between sections < 40°C20°C<
< 8°C
![Page 22: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/22.jpg)
22
Component Level Analysis
▪ Criteria 3: Pump Power
– One Pass requires less pump power
than Two Pass for the same flow rate
▪ Two Pass has smaller pipe diameter
and longer channel
![Page 23: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/23.jpg)
23
Component Level Analysis
▪ Criteria 3: Pump Power
– One Pass requires less pump power
than Two Pass for the same flow rate
▪ Two Pass has smaller pipe diameter
and longer channel
▪ Test shows advantages of designs
▪ Now test system in vehicle
– Control system, rest of physical system
– See which criteria is most important< 40°C20°C<
![Page 24: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/24.jpg)
24
Agenda
▪ Importance of Battery Cooling
▪ Exploring Battery Cooling Network Designs
▪ Integration in Vehicle Model
▪ Evaluation of Design in Full Vehicle Tests
![Page 25: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/25.jpg)
25
Electric Vehicle Model
![Page 26: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/26.jpg)
26
Electric Vehicle Model
▪ Battery Electric vehicle
▪ 3-Motor Architecture
– Rear: 40 kW Motor (2x)
– Front: 60 kW Motor
![Page 27: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/27.jpg)
27
Full Vehicle Test
▪ Integrate into Reference Application
from Powertrain Blockset
– Baseline model provides architecture
– Extend to 3 motor system
▪ Use Model-Based Design to
– Assess performance including
fuel economy and acceleration
– Develop control algorithms
– Deploy to hardware
Vehicle
Vehicle-level Controllers
DriverManeuver/Drive Cycle
Environment
![Page 28: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/28.jpg)
28
Powertrain Blockset
Library of blocks Pre-built reference applications
![Page 29: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/29.jpg)
29
Agenda
▪ Importance of Battery Cooling
▪ Exploring Battery Cooling Network Designs
▪ Integration in Vehicle Model
▪ Evaluation of Design in Full Vehicle Tests
![Page 30: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/30.jpg)
30
Scenario Testing
▪ 342 simulations:
– 2 cooling networks
– 57 drive cycles
– 3 temperatures: 0/20/40 °C
▪ Criteria
– Temperature range: 20-40 °C
– Temperature gradient: <8 °C
– Total cooling energy
▪ Accelerate testing
– Parallel Computing Toolbox
![Page 31: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/31.jpg)
31
Summary of Results
▪ Observations from 342 tests
– Two Pass has lower temperature difference
Less cell imbalance, better battery life
– One Pass has lower energy consumption:
Better fuel economy for same maximum temperature
Results from One Drive Cycle
![Page 32: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/32.jpg)
32
Key Takeaways
▪ Battery cooling network design
requires component level analysis
and tests within a full-vehicle simulation
▪ Integrating fluid, thermal, electrical,
and mechanical domains is key
to assessing system-level performance
▪ Rapid simulations covering a wide range
of drive cycles and ambient conditions
are needed to evaluate design criteria
Simscape
Model
Powertrain Blockset Model
Lumped Parameter Network
CFD and FEA
Computation
Time
Model Fidelity
Spreadsheet
![Page 33: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/33.jpg)
33
Products Used
▪ Battery Cooling Network Simscape, Simscape Fluids
▪ Electrical Network Simscape Electrical
Simscape Driveline
▪ Vehicle and Environment Powertrain Blockset
▪ Testing Simulink Test
Parallel Computing Toolbox
![Page 34: Virtual Vehicle Application: Battery Cooling Network Study · 3 Key Takeaways Battery cooling network design requires component level analysis and tests within a full-vehicle simulation](https://reader033.vdocuments.mx/reader033/viewer/2022050506/5f976da4f5bcdc37692aca28/html5/thumbnails/34.jpg)
34
Which tasks shown in this presentation
are most interesting to you?
Please contact us with questions
Speaker Image [email protected]
a Battery Modeling
b Cooling System Modeling
c Electrical Network Modeling
d Full Vehicle Simulation
e Parameter Sweeps and Results Analysis