RadTherm Battery ModelingUGM 2011- Antti Jussila

22Confidential

AgendaAgenda

Battery Design Issues Physical Description Model Description

Approach Implementation

Model Inputs Development Status

3

RadTherm Battery SimulationRadTherm Battery Simulation

3

1D System Tools Concept 1D Component Models Long Transients solved quickly

RadTherm Simple 1D, 2D or Detailed 3D Geometry Radiation, Conduction, Convection Long Transient Rapid Trade Studies

CFD Highly Detailed 3D Geometry Complex Flow Steady State or Short Transient

Optimization

Geo

met

ryD

etai

l

Simulated Time

1D SystemTools

Meshing & FEA

CFD RadTherm

CAE Coupling & Integration1D Tools – AMESim, Matlab, …CFD – Fluent, PowerFLOW, STAR-CCM+, …Optimization – Isight, modeFRONTIER, …Meshing – ANSA, Hypermesh, PowerDELTA,FEA – ABAQUS, NASTRAN

Confidential

44Confidential

Thermal IssuesThermal Issues

Battery Cells, Packs & Systems Cells – uniform current, temperature distribution Packs – uniform cooling, environmental effects Systems – Cell balance, SOC estimation

Cooling Strategies Air (dedicated or shared) Liquid (dedicated or shared) Heat sink, cooling plates, heat pipe, ….

Hybrid & Electric Vehicle Batteries Performance Lifetime & Durability Safety

55Confidential

Thermal ScenariosThermal Scenarios

Transient Drive Cycles Hot Soak Preheating for Cold Start Load Balancing Abuse Tolerance Cooling System Faults Thermal Runaway t = 120

min.cell 1cell 4

0.010.020.030.040.050.060.070.080.090.0

0 60 120 180 240 300 360 420 480 540 600

Spee

d(m

ph)

Time (seconds)

US 06 Drive Cycle

66

Battery Modeling DescriptionBattery Modeling Description

Effects of tab size/location, cell size,collector thickness: Voltage gradients on collector plates Non-uniform current density Non-uniform ion depletion Non-uniform heating

Electrical load

Effects of pack configuration, flowdistribution, packaging: Non-uniform cell behavior Non-uniform cooling

Effects of electrical load on total thermal load Charge/discharge profile

Cell Model:Examine local variations

Pack Model: examine variations incell-to-cell average behavior

Confidential

7

idischarge

7

Battery Model Implemented in RadThermBattery Model Implemented in RadTherm

Thermal problem Description of temperature distribution on

electrodes

Confidential

Cathode

Anode

Vanode = 0

2Vn = J / (sntn)

J = Y(Vp – Vn – U)

2Vp = -J / (sptp)

sp ( Vp Vp ) +J [ Eoc - ( Vp-Vn ) – T ( dEoc/dT ) ] / t +sn ( Vn Vn )

q’’’ =

Electrical problem Description of current density through electrodes Description of voltage distribution on collector

plates RadTherm solves thermal analog:

Thermal Conductivity = Electrical Conductivity Temperature = Voltage Heat Flux = Current Density

Two problems coupled via joule heating Calculated in Electrical Imposed on Thermal

88

Battery Model Setup in RadThermBattery Model Setup in RadTherm

Thermal problem Cell modeled as multi-layer part

Thermal properties of electrodes Convection on front, back face Optional edge conduction Hook function imposes Joule heating derived from electrical model

Confidential

Electrical problem Anode, cathode collector plates modeled as standard

parts k = electrical conductivity Hook function imposes a ‘heat flux’ equivalent to the current

density between electrodes

Anode tab is assigned a value of 0 volts (ref voltage) ‘Heat’ equal to discharge current is imposed on cathode

tab ‘Temperature’ solution is voltage distribution

So far, coupling is only one-way Electrical affects thermal Future development will add two-way coupling

99

Battery Cell Model ValidationBattery Cell Model ValidationCell data

Measured & Simulated discharge curves

2D Voltage Distribution

2D Temperature Distribution

Time (sec) DoDAvg Cathode Voltage(mV)

Anode Voltage(mV)

Cell CurrentDensity (A/m^2)

Uniformity Index

0 0 4.05E+03 -6.10E-06 -5.38E+02 060 0.016 4.03E+03 -6.10E-06 -5.24E+02 0.068

120 0.032 4.00E+03 -6.10E-06 -5.37E+02 0.057180 0.048 3.98E+03 -6.10E-06 -5.25E+02 0.048240 0.064 3.95E+03 -6.10E-06 -5.36E+02 0.042300 0.08 3.93E+03 -6.10E-06 -5.26E+02 0.037360 0.096 3.91E+03 -6.10E-06 -5.36E+02 0.033420 0.113 3.89E+03 -6.10E-06 -5.27E+02 0.03

Confidential

3

3.2

3.4

3.6

3.8

4

4.2

0 0.2 0.4 0.6 0.8 1DoD

Ter

min

alV

olt

age

(V)

0.5C (measured)1.0C (measured)1.5C (measured)

3

3.2

3.4

3.6

3.8

4

4.2

0 0.2 0.4 0.6 0.8 1

DoD

Ter

min

alV

olt

age

(V)

0.5C (measured)1.0C (measured)1.5C (measured)0.5C (RadTherm)1.0C (RadTherm)1.5C (RadTherm)

1010

Ajou Model* Provides Current Density ThroughAjou Model* Provides Current Density ThroughElectrodes as Function of Voltage DifferenceElectrodes as Function of Voltage Difference

Empirical model Current Density through Electrodes

Local current density = function of localpotential difference across cell J = Y ( Vp – Vn – U ) Vp = potential on cathode Vn = potential on anode U = fitting parameter ( open cell voltage)

U = aiDoDi ; i=0,1,2,... Y = effective conductivity of cell

Y = biDoDi ; i=0,1,2,... DoD = Depth of Discharge

DoD = Jdt / QT QT = theoretical capacity per unit area DoD varies across electrode area

Confidential* Also known as NTG model

1111

Required Battery Test dataRequired Battery Test data

Data needed to calculate Ajou coefficients Theoretical cell capacity Electrode area Discharge current – 1c, 2c, 4c Voltage data during discharge, correlated to

discharge current sufficient to resolve curve at beginning of discharge,

end of discharge

Confidential

1212

Battery Discharge Data to Coefficients

Obtain terminal voltage versusDoD at different values ofcurrent density:

Confidential

CrossCross--plot terminal voltageplot terminal voltageversus current density atversus current density atdifferent values of DoD:different values of DoD:

3

3.2

3.4

3.6

3.8

4

4.2

-140 -120 -100 -80 -60 -40 -20 0

Current Density (A/m^2)

Ter

min

alV

olt

age

(V)

DoD=0.05DoD=0.10DoD=0.20DoD=0.30DoD=0.40DoD=0.50DoD=0.60DoD=0.70DoD=0.80DoD=0.90

1313

U and Y determined from line fits to data

Confidential

Slope of line is 1/YDoD Y

0.02 330.15010.1 302.84150.2 312.83860.4 364.85620.6 372.46260.8 333.14410.9 303.0941

0

50

100

150

200

250

300

350

400

450

500

0.00 0.20 0.40 0.60 0.80 1.00

DoD

Y(S

/m^2

)

b_0 344.7134b_1 -690.134b_2 3656.48b_3 -5594.76b_4 2592.827

Y = 344.7 - 690.1DoD + 3656DoD2 -5595DoD3 + 2593DoD4

U = 4.127 – 1.654DoD + 1.908DoD2 -1.0193DoD3

3

3.2

3.4

3.6

3.8

4

4.2

0.00 0.20 0.40 0.60 0.80 1.00

DoD

U(V

)

a_0 4.126966a_1 -1.65359a_2 1.908175a_3 -1.01914DoD U

0.02 4.1041450.1 3.9763910.2 3.8721910.4 3.7024180.6 3.598930.8 3.5109450.9 3.431629

Intercept of line is U

1414ThermoAnalytics Proprietary

Next StepsNext Steps--Drive CycleDrive Cycle -- US06US06

Transient Current and Voltageinputs

1515

Data input from Excel spreadsheet or1D system tool

Current input, Temperature output

Transient data inputTransient data input--demo exampledemo example