Development and Application of Liquid-Cooled Lithium-Ion Battery Pack Thermal Model

Download as pdf or txt
Download as pdf or txt
You are on page 1of 14

Development and Application of Liquid-cooled Lithium-ion Battery Pack

Thermal Model
Model based approach by using GT-SUITE
Yifan (Flora) Zhou Meng Li
DEP DEP

Fan He Wei Tao


Optimal FCA US LLC

Xinran Tao
Optimal

GT’17 - Plymouth, MI
November, 2017 ©2017 FCA US LLC
Contents

• Background and motivation


• Battery electric-thermal modeling
• Application in production
• Conclusion
• Acknowledgement

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 2


Motivation: Battery Thermal Modeling

• Battery cell temperature and hot spots can have a critical impact on cell, pack and battery system
performance.
• Proper thermal management prevents:
• Thermal runaway
• Low fuel economy
• Reduced battery life
• Poor performance
• High-fidelity battery pack and cooling system model is important for accurate control and proper thermal
management.
ΔTemp
Hot spot
ΔTemp

Battery Cooling system


• Pump sizing
• Flow, heater and AC control
• Warm-up strategy
• …
North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 3
Single Cell Modeling Concept

• Diagram: Battery pack and cooling system


• Every single cell GT model includes electrical and thermal sub-models Vehicle
• This model is for an indirect liquid-cooling battery thermal system speed

Single Cell Cooling Fin Power


demand

Cell temperature
Heat generation
Battery:
Battery Electrical

Thermal Pad Thermal Pad Battery:


Cooling Plate Cooling Plate Thermal

Heat removal
by cooling
system
Cooling System

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 4


Battery Model Capturing Current and Heat Generation Non-uniformity

 To resolve current, heat generation and temperature non-uniformity within one cell, a detailed electrical
model consisting of cathode, anode and sub-cell electrical networks is required.
 A Single cell model consisting of 3 sub-cells (in 1x3 configuration) is shown below

Rci: electrical resistances on cathode


current collector
Rin: internal resistance of sub-cell
OCV: open circuit voltage of sub-cell\
H: Cell height
W: Cell width

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 5


Battery Cell Thermal Modeling
• A battery cell thermal model template has
been developed.

• Battery thermal behavior prediction


Battery cell model offers transient thermal Battery cell electric sub model for
current density/Heat generation
behavior prediction, given variable distribution calculation
charging/discharging cycles and different
coolant flow rates and temperatures
Qin
• Battery temperature distribution prediction
Battery cell thermal sub model for
Battery pack Max/Min temperature prediction, temperature distribution estimation
as well as coolant temperature change can be
achieved, by using simulation from knowledge Qout
of specific boundary conditions.
Battery cell cooling Battery cell cooling
--- natural pad/coolant flow
convection thermal mass for
heat removal rate
Battery estimation

Thermal Pad
Cooling Plate
North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 6
System Model Integration
• Assembly from previous slide battery cell model is packaged as a GT compound template.
• The battery cell temperature, coolant temperature and SOC can directly link to ICOS for system level
co-simulation.

Battery Pack Battery Cell

Cooling Pipe

ICOS: GT’s co-simulation template -- UserCodeCControls


North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 7
Heat Transfer Coefficient

• Heat transfer coefficient (HTC) at cooling pad on the coolant side is calculated through Dittus-Boelter
Correlation.
Dittus-Boelter Correlation: h: Heat transfer coefficient
d: Hydraulic diameter
k: Thermal conductivity
𝑘 𝑗 ∗ 𝑑 0.8 𝜇 ∗ 𝑐𝑝 0.33 μ: Fluid viscosity
ℎ = 0.23 ∗ ∗( ) ∗( )
𝑑 𝜇 𝑘 j: Mass flux
cp: Heat capacity of fluid
• Effective convection thermal resistance of cooling plate derived from HTC as a transient function of
coolant flow rate

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 8


Simulation Assumptions
• By neglecting electric work done in the electrolyte and the reversible entropic loss, which can be
neutralized during charging/discharging reaction, heat generation in one battery cell can be considered
to be mainly contributed by Joule losses calculated as
Q= 𝑅𝐼 2 [watt]
In which, I: Current, R: Internal resistance, Q: Heat generation rate

• Coolant inlet temperature and ambient temperature are set to be 18°C


• Battery cell divided into 3 uniform thermal masses. Inside each mass the heat generation is estimated
as a function of local current density and equivalent resistance, arranged as shown with bottom part
connected to cooling pad

Cooling surface

• Different coolant volume flow rates are investigated


• Define the transient current profile
• Convective heat transfer coefficient between battery and air is 5W/m2K

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 9


Simulation Cases Boundary Conditions

Cas Current Coolant Flow Rate Coolant Ambient T Battery Cell Initial SOC
e Profile (L/min) Inlet T (°C) (°C) Initial T (°C)
1 shown as Constant Flow
Current Profile 1C 18 25 25 0.9
2 Profile shown as
Coolant Flow

8,000s

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 10


Battery Pack Results Correlation

Case Current Coolant Flow Rate Coolant Inlet T Ambient T Battery Cell Initial T Initial
Profile (°C) (°C) (°C) SOC
1 1C Constant Flow 18 25 25 0.9
Heat generation from battery
Convection cooling &
charge & discharge
Cooling system effect
• Constant coolant mass flow rate
applied for battery temperature
stabilization

• Input electric current drops to 0A


at 6,000s, without heat
generation, the temperature
decreases very fast
6,000s 10,000s
• A good correlation between test
data and GT simulation results
achieved with an error smaller
than 1%. Max T @6,000s (°C) Coolant outlet T @10,000s (°C)
Error 0.9% 0.05%
North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 11
Battery Pack Results Correlation

Case Current Coolant Flow Rate Coolant Inlet T Ambient T Battery Cell Initial T Initial
Profile (°C) (°C) (°C) SOC
2 1C Coolant Flow 3 18 25 25 0.9
Heat generation from
battery charge & discharge
Convection cooling
• For the first 6000s, with no coolant
flow applied, cell temperature Convection cooling &
increases Cooling system effect

• Cell temperature drops slowly due


to air convection without heat
generation (current is 0A) during
6,000-8,000 sec of the simulation

• Coolant started flowing after 8,000


sec, the battery temperature 6,000s 8,000s 10,000s

decreases dramatically.
Coolant flow begins

Max T @6000s (°C) Coolant outlet T @10,000s (°C)

Error -3.2% 1.1%


North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 12
Summary

• GT battery cell model template was applied to a battery thermal model for transient heat
generation rate calculation, battery cell temperature distribution prediction and overall cooling
system performance evaluation

• Integrated battery pack thermal model, developed with proposed single cell model offers
accurate estimation of hottest individual cell temperature as well as battery pack coolant
temperature change at outlet

• Proposed GT template-based battery pack model provided a fast simulation of thermal


behavior and is convenient for integration into the whole vehicle level simulation

• Future work includes model based thermal management control strategy development and
experimental validation for improved simulation accuracy

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 13


Acknowledgement

The authors gratefully appreciate support from

• Oliver Gross
• Andrew Best
• Craig Linguist
• Pawel Malysz
• Joe Wimmer
• Brad Holcomb

North American GT Conference 2017 ©2017 FCA US LLC 11/06/2017 14

You might also like