Improving Battery Thermal Management Design using Six

Sigma Processby

Andreas Vlahinos, Kenneth Kelly,

John Rugh & Ahmad Pesaran

National Renewable Energy Lab

Golden, CO

Outline

• Project Objectives & Approach• Background Robust Design Techniques• Prismatic NiMH Battery Thermal

Analysis– Steady State Variation – Transient

• Summary and Future Plans

Objective and Approach

• The overall objective is apply variation analysis and Six Sigma design process to HEV battery thermal analysis– Better thermal performance– Longer battery life

• Approach: – Evaluate thermal management effectiveness of

cylindrical and prismatic batteries with air or liquid cooling for HEV or Fuel Cell vehicles using FreedomCAR 40 kW power profile

– Perform variation analysis and six sigma

Geometry and Load Cases

Two Battery Geometries:1. Prismatic 2. Cylindrical

Two Coolants1. Air Cooled2. Liquid Cooled

Two Ambient Load Cases:1. Buffalo, NY (-20°C amb) 2. Palm Springs (60°C amb)

Traditional Deterministic Design Approach

Accounts for uncertainties through the use of empiricalSafety factors:

– Are derived based on past experience– Do not guarantee safety or satisfactory performance– Do not provide sufficient information to achieve optimal use of

available resources

2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 00

0 . 0 0 5

0 . 0 1

0 . 0 1 5

0 . 0 2

0 . 0 2 5

0 . 0 3

0 . 0 3 5

SupplyResistance Capacity

DemandLoad

Requirement

Six-sigma Design

– Identifying & qualifying causes of variation

– Centering performance on specification target

– Achieving Sigma level robustness on the key product performance characteristics with respect to the quantified variation

0.002

3.4

317311

45500

2700

63

0.57

308538

66807

6210

233

0

0

0

1

10

100

1,000

10,000

100,000

1,000,0000 1 2 3 4 5 6 7

Sigma Quality Level

Def

ects

(pa

rts

per m

illio

n)

Centered1.5 sigma shift

Robust Optimization

0.8 0 .85 0 .9 0 .95 1 1.05 1.1 1.1 5 1.20

2

4

6

8

1 0

1 2

1 4

Th ic kn e s s t (m m )

Pro

babi

lity

Den

sity

Th ic kn e s s Va ria tio n (No rm al) w ith µ t=1 .1 0 m m & σ t=0 .0 3 3 m m

0.8 0 .85 0 .9 0 .95 1 1.05 1.1 1.1 5 1.20

2

4

6

8

1 0

1 2

1 4

1 6

Th ic kn e s s t (m m )

Pro

babi

lity

Den

sity

Th ic kn e s s Va ria tio n (No rm a l) w ith µ t=0 .8 8 m m & σ t=0 .0 2 6 4 m m

0 10 20 30 40 50 60 700

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

Lo ad (N)

Rel

ativ

e Fr

eque

ncy

His to g ram o f Lo ad Variatio n (Lo g n o rmal)

-150 -100 -50 0 50 100 150 200

0.01

0.02

0.03

0.04

0.05

0.06

Rel

ativ

e Fr

eque

ncy

µt=2.00 mm

Histograms of Performance Function G for t=0.85, 1.0 and 2.0 mm

0.8 1 1.2 1.4 1.6 1.8 2-250

-200

-150

-100

-50

0

50

100

150

200

Me an Valu e o f Th ic kn e s s µ t (mm)

Mea

n V

alue

of G

- n σ

G (MP

a)

µG - 1 σG µG - 2 σG µG - 3 σG µG - 4 σG µG - 5 σG µG - 6 σG

0

Performance Function G MPa

Robust Optimizationreusable workflow template

Ford Motor CompanySAE – IEBEC 2001

0 10 20 30 40 50 60 700

0.005

0.01

0 .015

0.02

0 .025

0.03

0 .035

0.04

0 .045

0.05

Lo ad (N)

Rel

ativ

e F

requ

ency

His to g ram o f Lo ad Variatio n (Lo g no rmal)

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=1 .10 mm & σ t=0 .03 3 mm

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

16

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=0 .88 mm & σ t=0 .02 64 mm

Robust Optimizationreusable workflow template

Ford Motor CompanySAE – IEBEC 2002

0 10 20 30 40 50 60 700

0.005

0.01

0 .015

0.02

0 .025

0.03

0 .035

0.04

0 .045

0.05

Lo ad (N)

Rel

ativ

e F

requ

ency

His to g ram o f Lo ad Variatio n (Lo g no rmal)

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=1 .10 mm & σ t=0 .03 3 mm

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

16

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=0 .88 mm & σ t=0 .02 64 mm

Robust Optimizationreusable workflow template

Daimler ChryslerSAE Powertrain Conference

0 10 20 30 40 50 60 700

0.005

0.01

0 .015

0.02

0 .025

0.03

0 .035

0.04

0 .045

0.05

Lo ad (N)

Rel

ativ

e F

requ

ency

His to g ram o f Lo ad Variatio n (Lo g no rmal)

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=1 .10 mm & σ t=0 .03 3 mm

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

16

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=0 .88 mm & σ t=0 .02 64 mm

Robust Optimizationreusable workflow template

0 10 20 30 40 50 60 700

0.005

0.01

0 .015

0.02

0 .025

0.03

0 .035

0.04

0 .045

0.05

Lo ad (N)

Rel

ativ

e F

requ

ency

His to g ram o f Lo ad Variatio n (Lo g no rmal)

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=1 .10 mm & σ t=0 .03 3 mm

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

16

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=0 .88 mm & σ t=0 .02 64 mm

Plug PowerASME / RIT Fuel Cell Conference

Robust Optimizationreusable workflow template

USABC / FordAmerican Society of Quality

0 10 20 30 40 50 60 700

0.005

0.01

0 .015

0.02

0 .025

0.03

0 .035

0.04

0 .045

0.05

Lo ad (N)

Rel

ativ

e F

requ

ency

His to g ram o f Lo ad Variatio n (Lo g no rmal)

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=1 .10 mm & σ t=0 .03 3 mm

0.8 0.85 0.9 0.9 5 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14

16

Th ic kn e s s t (mm )

Pro

babi

lity

Den

sity

Th ic kn e s s Va riatio n (No rm al) w ith µ t=0 .88 mm & σ t=0 .02 64 mm

Reusable Process TemplateOptimization Loop PDS Loop

PDMµtga σtgapµR, σRµFrate, σFrate

Sampling Technique:Box-Behnken Matrix design

Regression analysis technique:Forward-stepwise-regression

µTmax, σTmaxµdTσdT

tgapRFrate

Parametric Deterministic

CAD/FEA Model

TmaxdTdP

σ Level for Tmax targetσ Level for dT targetσ Level for dP target

Sampling Technique:LHS, CCM or D-Optimal

Regression analysis technique:Forward-stepwise-regression

Optimization Method:Sequential Unconstrained

Are σQuality Levels

Acceptable?

no

Steady State Thermal Analysis with Input

Parameter Variations

Inputs with Variation

• Gap Thickness

• Cell Resistance

• Flow Rate• Six input

parameters:1. µtgap2. σtgap3. µR4. σR5. µFrate6. σFrate

Model Outputsmax temperaturedifferential temperaturepressure drop

Six output parameters:1. µTmax

2. µdT

3. µdP

4. σTmax

5. σdT

6. σdPThree Upper SpecificationLimits (USL)

Temperature Differential and Sigma Quality Levels

Sigma Quality Level

Design Space with σ Quality Regions Tmax

Design Space with σ-Quality Regions dT

Design Space with σ-Quality Regions dP

Design Space with σ Quality Regions All Parameters

Sigma Quality Levels Versus Mean Value

Sensitivity Analysis– The flow rate has the

most impact on the maximum temperature

– All three input design variables have about equal effect on the temperature differential

– The internal battery resistance has no effect on the pressure drop.

Transient Thermal Response of

Prismatic Batteries

FreedomCAR 40-kW Baseline Power Assist & Heat Generation Profiles

Inlet Temperatures for Air and Liquid Cooling

Inlet, Outlet and Core Temperature for Palm Springs with High Air Flow Rate

Core Temperature for Palm Springs with High Air Flow Rate

Core Differential Temperature for Palm Springs with High Air Flow Rate

Average Core Temperature(Palm Springs)

Distribution of Max Core Temperature

Average Core Temperature(Buffalo)

Summary• Demonstrated a re-usable process for including

statistical variation of input parameters for battery thermal analysis

• Initial analysis with variation shows:– Tmax is most difficult criterion to achieve with the

given design constraints and assumptions– Effect of conflicting design constraints on sigma

quality level• Completed first round of transient thermal analyses on

prismatic design• Initial transient results show

– the importance of including transient analysis– liquid cooling is more effective, but pressure drop

higher– transient cooling and warm up time of the heat

transfer fluid needs to be considered.

Future plans

1. Introduce feedback control on the fan1. Fan on–off, speed levels, etc2. Evaluate the effectives of various control systems on

thermal performance.2. Find the effect of power pulses in the load cycle on

thermal transient .3. Obtain a non-uniform heat generation profile from

published information or other test data (thermal imaging).

4. Modify duty-cycle to include appropriate diurnal ambient and load conditions

5. Perform transient thermal analysis on cylindrical battery pack

Acknowledgments

This research effort was funded by the Department of Energy (DOE), Office of the FreedomCAR and Vehicle Technology.

The authors would like to express their appreciation to:– Robert Kost, team leader of the FreedomCAR and

Vehicle Technology office– Tien Duong, Technology Manger of

Electrochemical Energy Storage program and – Ted J. Miller of Ford Motor Company and

FreedomCAR Battery Tech Team Chairman– Bruce Bryant, of Ford Motor Company