ultracapacitor applications and evaluation for hybrid ... · ultracapacitor applications and...

27
Ultracapacitor Applications and Evaluation for Hybrid Electric Vehicles Ahmad Pesaran ([email protected] ) Jeff Gonder, Matt Keyser National Renewable Energy Laboratory Presented at the 7th Annual Advanced Capacitor World Summit Conference NREL/PR-540-45596 NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Upload: vutuong

Post on 14-Apr-2018

226 views

Category:

Documents


4 download

TRANSCRIPT

Ultracapacitor Applications and Evaluation for Hybrid Electric Vehicles

Ahmad Pesaran([email protected])

Jeff Gonder, Matt KeyserNational Renewable Energy Laboratory

Presented at the 7th Annual Advanced Capacitor World Summit ConferenceNREL/PR-540-45596

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

2

Discussion Points

• Discussion of batteries vs. ultracapacitors for advanced vehicles

• Simulation results of HEV fuel economy impact from reducing the storage system’s energy window

• 15%-30% HEV fuel economy improvements with 50-100 Wh ultracapacitors

• Evaluation of lithium ion capacitors for HEV applications

• Thermal evaluation of a high-voltage ultracapacitor module for start-stop applications

3

Strong Attributes of Ultracapacitors Potential Specific UseHigh specific power and efficiency Engine assist

Efficient and fast charge acceptance Regen capture

Low resistance Lower cooling needs (less expensive)Quick response (short time constant) Supporting engine transients

Long anticipated calendar and cycle life Fewer replacements (less expensive)

High specific power at low temperatures (cold starts) Smaller size and less expensive

Weak Attributes of Ultracapacitors Specific UseLow specific energy Limited “durations” for power drawHigh self-discharge Loss of functionality and balance at start

Quick voltage variation More difficult to controlLow energy density Limited time for running auxiliaries at idle

High cost per unit energy Too expensive currently

The best use for Ucaps are strategies that make engines operate more efficiently (idle off, load leveling), frequent use capturing regen energy, and start-stop.

Strengths and Weaknesses of Ultracapacitors

4

A Couple of Thoughts

• Taking advantage of an ultracapacitor’s strengths while minimizing the impact of its weaknesses to make its “value” competitive with batteries

• It should be for a specific application to show “value” in terms of “life-cycle cost”

— Fuel economy— Replacement cost— Life— Durability and reliability— Quality— Functionality

5

Event How Much Energy Needed

Assist:20/30 kW constant power for 15/10 s 83.3 Wh

Accessory: 3 kW constant draw for 1 minute 50

Accessory: 1 kW constant draw for 1 minute 16.7

2% Grade going 35 mph for 1 minute Ŧ 70 Wh4% Grade going 35 mph for 1 minute Ŧ 170 Wh US06 Driving Cycle * 155 Wh

UDDS Driving Cycle * 80 Wh

Ucap Is Energy Limited!How Much Energy Is Needed for Various Events?

Prius has a 1.4 kWh NiMH battery but capacity is for life margin and warranty.Vue mild hybrid has a 0.6 kWh NiMH battery.

•Total Energy (at wheels) calculated for 1520 kg vehicle (regen); 50% of energy in the cycle’s largest deceleration eventCold-start capability is expected to dictate the size of batteries, but not the case for Ucap.

Ŧ Note: Engine provides propulsion up a grade, the estimate is for capturing regen to hold a 1520 kg vehicle speed going down a grade.

6

Potential Use of Ultracapacitors in Light-Duty Electric-Drive Vehicles

NiMH and Li-ion: YesUcap: LikelyUcap + VRLA: Possible

Micro Hybrids (12 V-42 V: Start-Stop, Launch Assist)

Li-ion: YesUcaps + high energy Li-ion : Possible

Plug-in HEV (EV)

NiMH and Li-ion: Yes Ucaps: Likely if Fuel Cell is not downsizedUcaps + (NiMH or Li-Ion): Possible

Fuel Cell Hybrids

NiMH and Li-ion: YesUcaps: Possible Ucaps + (NiMH or Li-Ion): Possible

Full Hybrids (150 V-350 V: Power Assist HEV)

NiMH and Li-ion: YesUcaps: Likely if engine is not downsizedUcaps + VRLA: Possible

Mild Hybrids (42 V-150 V: Micro HEV Function + Regen)

15-25 Wh

25-70 Wh

60-150 Wh

60-150 Wh

5-20 kWh(50-90 Wh*)

Min energy needed

* Energy for a Ucap in combination withLi-Ion

7

Analyzing the Impact of Energy Window on Power-Assist HEVs

• Motivation: Investigate the relation between in-use energy window and fuel economy (a request from USABC/FreedomCAR)

• Approach: Simulate a midsize sedan with different component power levels and control settings for different drive cycles using PSAT

Midsize Car Assumptions

Mass = 1675 kgEngine = 90 kWRESS/Motor = 30 kWElec accessories = 500 WMech accessories = 230 W

FA = 2.27 m2

CD = 0.30Crr1 = 0.008Crr2 = 0.00012

Simulated different ES energy content cases with the otherwise constant platform values

Constant 30 kW power changing P/E ratio

Smallest ES energy

Largest ES energy

Wh

Upper threshold

Target level

Lower threshold

Constant SOC-based controls (charge sustaining)

Changing Wh control window toleranceSource: J. Gonder, Presentation to USABC, July 19, 2007

8

Definition of ES Energy Window Use (for a drive cycle or event)

Energy out for electric launch/assist

Cum

ulat

ive

RES

S W

h to

veh

icle

RESS use indicated by slope of energy line

Energy return from charging/regen

“Energy window” defined by (max – min) for the

particular cycle

(not “target window” from control strategy)

Charge sustaining over cycle

(no net energy use)

Energy Window Used ≤ Available EnergySource: J. Gonder, Presentation to USABC, July 19, 2007

9

US06 Cycle

0 500 1000 1500-20

0

20

40

60

80

time (s)

Vehicle speed (mph)Cycle grade (%)

• Mean power during:Propulsion = 21 kWDeceleration = -17 kW

• No grade

“NREL to Genesee Cycle”

Aggressive driving

Up and down, foothills driving

0 200 400 6000

20

40

60

80

100

time

Vehicle speed (mph)

time (s)

• Mean power during:Propulsion = 23 kWDeceleration = -12 kW

• Considerable grade

0 500 1000 15000

20

40

60

time (s)

Vehicle Speed (mph)

UDDS CycleMild urban driving

• Mean power during:Propulsion = 7 kWDeceleration = -5 kW

• No grade

p

Constant 30 kW power changing P/E ratio

Smallest ES energy

Largest ES energy

p

Constant 30 kW power changing P/E ratio

Smallest ES energy

Largest ES energy

Source: J. Gonder, Presentation to USABC, July 19, 2007

Three Cycles Simulated to Observe Energy Window and Fuel Use

10

On City Cycle (UDDS), Large Fuel Savings Result from Hybridization

6.50

7.50

8.50

9.50

10.50

11.50

12.50

0 100 200 300 400ESS Energy Window (Wh)

Fuel

Con

sum

ptio

n (L

/100

km

)

(mpg

refe

renc

e)

18.8

20.5

22.4

24.8

27.7

31.4

36.2

Conventional Vehicle

Smallest RESS case

Largest RESS case

21% decrease

18% decrease

RESS Energy Window (Wh)Source: J. Gonder, Presentation to USABC, July 19, 2007

11

6.50

7.50

8.50

9.50

10.50

11.50

12.50

0 50 100 150 200 250 300 350 400

ESS Energy Window (Wh)

Fuel

Con

sum

ptio

n (L

/100

km

)

(mpg

refe

renc

e)

18.8

20.5

22.4

24.8

27.7

31.4

36.2

Summary Results of ES Energy Window and Fuel Economy Simulations

UDDS Cycle

US06 Cycle

Foothills Driving

Same Vehicle (largest RESS)

Same Vehicle (smallest RESS)

Same Vehicle (conventional)

RESS Energy Window (Wh)

Source: J. Gonder, Presentation to USABC, July 19, 2007

12

3

4

5

6

7

8

9

10

11

0 100 200 300 400 500 600 700

Energy Window (Wh)

Fuel

Con

sum

ptio

n (L

/100

km)

Fuel Economy (m

pg)

58.81

47.0

39.20

33.60

29.40

26.14

23.52

21.3

78.41

Charge SustainingNot Charge SustainingUDDSUS 06HWFTMTSS Speeds

Prius Camry Escape Accord

All the charge sustaining (CS) tests use windows <200 Wh(for these vehicles and CS cycles)

Test data analysis seems to validate simulation finding of significant hybridization

benefit in the 50-150 Wh range

Vehicle Test Results: Battery Energy Use for Today’s HEVs under Various Drive Cycles

Prius and Escape Test Data: Tony Markel, NRELCamry and Accord Test Data: Mike Duoba, ANLTest Data Analysis: Jaehun Rhee and Jeff Gonder, NREL

Source: J. Gonder, Presentation to USABC, July 19, 2007

13

2007 Mild Hybrid Dyno Data* Analysis Indicates <50 Wh Energy Use for Typical

Driving—Already Reasonable Ucap RangeDriving Energy Analysis (UDDS cycle example)

Energy window

* Department of Energy-sponsored dynamometer testing

14

Mild and Power-Assist Hybrids with Ucaps

• It is possible to use ultracapacitors (with available energy of 50-150 Wh) in power-assist HEVs with modest fuel economy improvements

— However, acceleration and passing on grade performance considerations could be limiting factors

• 15%-30% HEV fuel economy improvements with 50-100 Wh ultracapacitors

• A project is underway on a vehicle to demonstrate Ucaps in mild hybrids

— To be discussed in future meetings

15

0

2

4

6

8

10

12

14

16

18

20

0 100 200 300 400 500 600time (s)

Vo

lta

ge

(V

)

Hybrid Pack – UC VoltageBattery-only – Voltage

Previous NREL Tests Have Shown That Combining Ultracapacitors Filters High Current Transients In Batteries

Source: M. Zolot (NREL Reports and 2003 Florida Capacitor Seminar)

Ultracapacitor module of 8 cells (up to 20V) and two 6.5Ah NiMH module of 14.4V (18V max). Ultracap module and battery pack were arranged in parallel to share the current load depending on internal impedance.

0

5

10

15

20

25

30

35

40

45

-40 to -110

-30 -20 -10 0 10 20 30 40 to100

Current (A)

g g g

% Freq Hyb_NiMH % Freq Bat_NiM

ChargeDischarge

Battery-only - Battery Current

Hybrid Pack - Battery Current

• Overall, batteries in the hybrid pack experienced no currents larger than ±40 A, while the batteries in traditional pack saw currents up to ±110 A.

• Up to 33% narrower battery SOC cycling range was observed in hybrid pack; this has the potential to increase battery life.

Parallel connection; no DC/DC converterMay not be practical to implement in vehicles.

16

Advantages/Disadvantages ofHybridizing Energy Storage (Ucap + Battery)

Advantages• Reduced battery currents• Reduced battery cycling range• Increased battery cycle/calendar life (to

what extent?)• Increased combined power and energy

capabilities• Lower cooling requirements• Better low-temperature performanceDisadvantages• Complex control strategy• Larger volume & mass• Need for electronics for each system• Increased energy storage cost• Unknown side effects if directly coupled• Any need for DC/DC converters adds

even more cost and complexity

+

UcapModule

+

BatteryModule

DC

DCDC

+EM14V

Energy Storage Subsystem

EE300-20

AC

Power

DC

DC

UUCap > UBatt

DC

DC

UUcap > UBatt+

UcapModule

+

BatteryModule

DC

DCDC

+EM14V

Energy Storage SubsystemEnergy Storage Subsystem

EE300-20

AC

Power

DC

DC

UUCap > UBatt

Power

DC

DC

Power

DC

DC

UUCap > UBattUUCap > UBatt

DC

DC

UUcap > UBattUUcap > UBatt

Source: Continental ISAD, “New Energy Storage Concept,” Proceedings of AABC-04

17

• JSR Micro contacted us to express interest in thermal characterization of their asymmetric capacitor

• JSR Micro claimed higher energy than C-C Ucaps with the same power capability

• We received 3 cells for characterization per USABC protocols

Thermal/Electrical Characterization ofJSR Micro Lithium Ion Capacitor (LIC)

Source: www.jmenergy.co.jp/en/product.html

18

JSR Micro LIC Cell Characteristics

Cell 1 0.205 2.669 5.5" x 4" x 0.330" 1.58Cell 2 0.205 2.669 5.5" x 4" x 0.330" 1.62Cell 3 0.205 2.672 5.5" x 4" x 0.330" 1.6

Impedance (mOhms)Cell Number

(#) Mass (kg)Voltage (Volts)

Dimensions (inches)

Nominal 2200 F14 Wh/kg3.8 V – 2.2 V Source: www.jmenergy.co.jp/en/product.html

19

Infrared Thermal Imaging

+ -

+

Cell #1 Cell #2 Cell #3

-

--

+

+

+

Temperatures: AmbientProfiles: 50C, 100C, and Geometric Cycle

20

Thermal Image and Thermal Lines of 3 LIC Cells – 100 A Discharge

21

Thermal Characterization in NREL Calorimeter Lithium Ion Capacitor 2200 F Cells

•Temperatures: +30˚C• Profiles: CC discharge cycles

0 0.5 1 1.5 2 2.5 3 3.5 4

Time (Hours)

Calorimeter Response to Constant Current Charge/Discharge

Hea

t Gen

erat

ion

Discharge - Exothermic

Charge - Endothermic

Increasing Discharge Current

Increasing Charge Current

22

Electrical Characterization: Lithium Ion Capacitor Cells

• C/1, 10 C, 100 C, and HPPC Testing

2

2.2

2.4

2.6

2.8

3

3.2

3.4

3.6

3.8

0 10 20 30 40 50 60 70 80 90 100DOD (%)

OC

V (V

olts

)

Energy: 14 Wh/kg

Power: 1500 W/kg

This asymmetric capacitor had high resistance; the next generation is claimed to be better.

2200 F cell 0 20 40 60 80 100 120Depth of Discharge (%)

Discharge Regen

Pow

er

HPPC Discharge/Regen Power

0.7

0.75

0.8

0.85

0.9

0.95

1

0 20 40 60 80 100 120Current (Amps)

Dis

char

ge C

apac

ity (A

h)

Cell #1 Cell #2 Cell #3

23

Expected Calendar Life of Typical Current EDLC TechnologyMuch Better Than Batteries if Stored at Low Voltages

70 C

60 C

50 C

40 C30 C

Clamped Voltage

Expe

cted

Cal

enda

r Life

(Yea

rs) Source: Anderman, 2004 Advanced

Automotive Battery Conference

24

Thermal Evaluation: High-Voltage Ultracap Module

• Tested as part of USABC deliverable• Eighteen (18) symmetric carbon-carbon ultracapacitors• Tested under realistic conditions and operation• Used different power profiles and chamber temperatures

5

0

5

0

5

0

0 1 2 3 4 5 6

Time (Hours)

Terminal Temperatures after One Hour Geometric Cycling

5 6 13 14

4 7 12 15

3 8 11 16

2 9 10 17

1 18

Balancing Board

5 6 13 14

4 7 12 15

3 8 11 16

2 9 10 17

1 18

Balancing Board

Balancing Board Affecting Terminal Temps

Heat from cells is conducted through the ends to the case and rejected through the top metal heat sink/fins.

25

Thermal Evaluation: High-Voltage Ultracap Module

• Continuous US06 cycling for two hours • Balancing board did a good job equalizing cells• Energy drain for balancing could be a concern

0

5

10

15

20

25

30

35

40

45

0 2 4 6 8 10 12 14 16 18 20Time (Hours)

Bal

ance

Boa

rd T

empe

ratu

re (

C)_

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Max

to M

in C

ell V

olta

ge (

Vol

ts)_

Balance Board Temp

Max to Min Voltage Difference

29

30

31

32

33

34

35

36

0 1 2 3 4 5 6Time (Hours)

Te

rmin

al T

em

pe

ratu

re (

C)

Cell Terminal #1 Cell Terminal #3 Cell Terminal #5 Cell Terminal #7 Cell Terminal #9Cell Terminal #11 Cell Terminal #13 Cell Terminal #14 Cell Terminal #16 Cell Terminal #18

Cell #1

Interior Cells

Exterior Cells

(next to balancing board)

Temperature difference less than 1.5°C except for Cell #1 which heated due to balancing board.

26

Concluding Remarks• Ultracapacitors provide opportunity for modest fuel savings in

hybrid cars— Idle-off: 5%-10% FE improvement and most likely to be implemented— Mild and full hybrid: 15%-25% FE improvement, possible— Plug-in hybrids: possible Ucap combined with batteries; cost??

• Competition from Li-ion is strong; ultracapacitors should provide “added value” to compete

— Low-temp performance— Longer cycle and calendar life

• Asymmetric capacitors such as lithium ion capacitors have potential if power and cost are improved

• Thermal issues are important and must be taken into account to achieve the desired performance and life

• Lower cost is the key for increased market growth in automotive• Micro and mild hybrids provide biggest opportunity for Ucaps in

the short term; will be accelerated by new CAFÉ mandates

27

Acknowledgements

• Support provided by FreedomCAR and Fuel Partnership in the Vehicle Technologies Program of the U.S. Department of Energy

— David Howell, Energy Storage Technology Manager

• Technical insight and support— Harshad Tataria, USABC/GM— Jim Banas, JSR Micro Inc.

nrel.gov/vehiclesandfuels/energystorage/publications.html