PHEV Energy Storage and Drive Cycle Impacts

7th Advanced Automotive Battery ConferenceLong Beach, California

May 17th, 2007

Tony Markel and Ahmad Pesaran(Tony_Markel@nrel.gov)

National Renewable Energy Laboratory

Supported by FreedomCAR and Vehicle Technologies Program

Office of Energy Efficiency and Renewable Energy U.S. Department of Energy

NREL/PR-540-42026

Outline

• Background - Summary of Previous Work• Key Messages of this Study • Real-World Drive Cycles• PHEV Recharge Options• Operational Impacts on

— Pulse Power — State of Charge

• Conclusions

Key Messages of this Study

• Petroleum Consumption— The fuel displacement benefits of PHEVs will be

influenced by the frequency of recharging events• Pulse Power Attributes

— PHEVs are likely to encounter long pulse power events during real-world duty cycles

— PHEV experiences similar power levels but much longer pulses than HEV

• State of Charge— Time at specified state of charge varies significantly

with platform and recharge scenario

Standard and Real-World Drive Cycles

• Standard drive cycles used for certification/comparison purposes,— UDDS, HWFET, US06, SC03— Japan-1015— NEDC

• These drive cycles are meant to be representative for test efficiency— Fuel economy labeling under revision and likely to be based on

broader set of cycles to address differences between labels and consumer experience

• Real-world driving patterns provides insight on in-use speed and acceleration characteristics

— PHEV recharge scenarios and grid impacts can be better analyzed with time of day information

Drive cycle Vehicle Simulations

Fuel EconomyFuel Cost

Power NeedEnergy Need

Vehicle attributes

Battery SizingVehicle Cost

• Key insights— Speed and acceleration distributions— Time of day usage for recharge analysis— Combined impact of speed and grade— Location and duration of stops for recharge opportunities

Real-World Drive Cycle Resources

• Driving/travel survey is ongoing in many cities (e.g., St. Louis)• Augmenting these surveys with GPS information from individual

vehicles provide details needed for simulation

• 1Hz data collected — Time of day— Speed— Altitude— Latitude— Longitude

Sample Real World Duty Cycle

0 5 10 15 20 25 300

20

40

60

80Sp

eed

(mph

)

Time (hr)

0 5 10 15 20 25 30 35 40 45 500

20

40

60

80

Spee

d (m

ph)

Distance (mi)

PHEVs Reduce Fuel Consumption By >50% On Real-World Driving Cycles

Vehicle in-use activity pattern and simulated fuel consumption

— In-use bars show morning, midday and evening usage peaks; at most 12% of vehicles in use at once

— Cumulative fuel consumption lines consider entire fleet using specified architecture

Assumptions• PHEVs begin fully charged and do not

charge until they finish driving for the day• Gasoline is $2.41/gallon and electricity is

$0.09/kWh for energy cost comparison (purchase price differences not included)

St. Louis Data Sample = 227 Vehicles

Four Potential Daily Recharge Strategies

Immediate End of Travel Day

3 ways to control a single daily charge Multiple charging events per day

Recharge Scenario Impacts on PHEV Petroleum Consumption Benefits

OpportunityCharge(PHEV20)

No Charge (PHEV20)

Opportunity charge: connect PHEV charger to grid any time that the vehicle is parked.

Base Case assumes one full charge per day

Outline

• Background - Summary of Previous Work• Key Messages of this Study • Real-World Drive Cycles• PHEV Recharge Options• Operational Impacts on

— Pulse Power — State of Charge

• Conclusions

Two Ways to Analyze Battery Power Profiles

• Power Pulse event— Start - first non-zero— End – next non-zero

• Attributes— Peak power and peak power duration— Energy equivalent average power and duration

• Provides detail on specific events

• Moving window approach — Integrate power profile over a specified window to

find net, positive only, and negative only equivalent powers

— Captures interaction between multiple events

Characteristics of an Individual Pulse Power Event

0 5 10 15 20 25 300

20

40

60Sp

eed

(mph

)

0 5 10 15 20 25 300

0.5

1

SOC

(--)

0 5 10 15 20 25 30-50

0

50

100

Pow

er (k

W)

Distance (mi)

Engine

Battery

0 5 10 15 20 25 300

20

40

60

Spee

d (m

ph)

0 5 10 15 20 25 300

0.5

1

SOC

(--)

0 5 10 15 20 25 30-50

0

50

100

Pow

er (k

W)

Distance (mi)

Engine

Battery

Characteristics of an Individual Pulse Power Event

Expanding window captures event interactions

Moving Window Analysis of ESS Power Profile Quantifies Interaction Between Individual Events

• Determined Energy Equivalent Pulse Power for Spectrum of Durations

• Moving Window and Individual Event pulse power the same when window duration equals event duration

0 50 100 150 200 250-30

-20

-10

0

10

20

30

40

50

60

Average Power Pulse Duration (s)

Max

imum

Ave

rage

Pow

er P

ulse

(kW

)

Max. Moving Window Power

Min. Moving Window PowerAvg. Power of Events

Peak Power of Events

Detailed Pulse Power Analysis of Real Travel Profile Identifies Most Challenging Events

Key Events

Short duration high power

Long duration moderate power

Multiple short events interact in mid duration range

Curve higher than pointsHighlights event interaction

Midsize Car PHEV20

w/AER on UDDS

Conv.HEV

Pulse Power Characteristics Depend on Operating Strategy

UDDS CDE

Total Event Duration

Pow

er

Time

Peak Pulse Power Duration

Energy Equiv. Pulse Power

EngineBattery

Charge depleting electric (CDE) is likely to have short high power events and moderate long duration energy equiv. events.

Conv.HEV

Pulse Power Characteristics Depend on Operating Strategy

Vehicles designed as CDE on UDDS are likely to operated as CDH on real-world duty cycles!

UDDS CDE

Total Event Duration

Pow

er

Time

Peak Pulse Power Duration

Energy Equiv. Pulse Power

EngineBattery

Charge depleting electric (CDE) is likely to have short high power events and moderate long duration energy equiv. events.

Charge depleting hybrid (CDH) will have lower but longer peak pulse and slightly lower energy equiv. pulse power requirements

Low CDH

Conv.HEV

Pulse Power Characteristics Depend on Operating Strategy

Vehicles designed as CDE on UDDS are likely to operated as CDH on real-world duty cycles!

UDDS CDE

Total Event Duration

CDH

Pow

er

Time

Peak Pulse Power Duration

Energy Equiv. Pulse Power

EngineBattery

Low CDH

Charge depleting electric (CDE) is likely to have short high power events and moderate long duration energy equiv. events.

Charge depleting hybrid (CDH) will have lower but longer peak pulse and slightly lower energy equiv. pulse power requirements

In CDH lower power case, the Peak and Energy Equiv. Pulse Powers may have similar level and duration

Pulse Power from Simulated PHEV Operation on 227 Real-World Travel Profiles

0 20 40 60 80 100 120 140 160 180 200-30

-20

-10

0

10

20

30

40

50

60

Pulse Power Duration (s)

Max

. Ene

rgy

Equi

v. P

ulse

Pow

er (k

W) 10th Percentile

50th Percentile90th Percentile

• Components sized for AER on UDDS (CDE) still encounter long duration energy equiv. power pulses

May be necessary to specify in this region too

Typically specifiedat 2S and 10S

Midsize Car PHEV20 w/AER on UDDS

Pulse Power Analysis Methods Can be Applied to Both Simulation Results and Test Data

0 50 100 150 200 250-30

-20

-10

0

10

20

30

40

50

60

Average Power Pulse Duration (s)

Max

imum

Ave

rage

Pow

er P

ulse

(kW

)

Max. Moving Window Power

Min. Moving Window PowerAvg. Power of Events

Peak Power of Events

PHEV Research Vehicle Urban Driving Data

Simulated PHEV20Vehicle Results

PHEV Time At SOC Impacted by Charging Scenario

0

10

20

30

40

50

60

70

80

90

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1State of Charge Zone

% o

f Day

at S

tate

of C

harg

e

HEVPHEV20-basePHEV20-opchgPHEV20-nochg

Will differences in Time at SOC affect battery

life?

Based on Simulation of 227 duty cycles from St. Louis

Conclusions

• Pulse Power Analysis Methods— Moving window allows evaluation of interaction of pulse power events

• Petroleum Consumption Relative to Conventional Fleet— PHEV20 with single daily charge saves about 50%— PHEV20 without charging similar to HEV (~35%)— PHEV20 with opportunity charge saves 75%

• Pulse Power Attributes— Real-world pulse power events have longer durations than standard

test cycles— PHEV similar power levels but much longer pulses than HEV— CDH peak power is lower but duration is longer than CDE— CDH energy equiv. power is slightly lower with duration same as CDE

• State of Charge— No charge leads to long periods of battery at low SOC— Single charge leads to mixture of high and low SOC operation— Multiple charges leads to more time at high SOC

Next Steps

• Use battery models representative in both short and long duration pulses

• Determine key aspects affecting battery life• Continue to use travel data to assess impacts of PHEV

technology, especially on batteries— Charge-depleting electric and charge-depleting hybrid

operating scenarios— PHEV10 scenario— Affect of ambient conditions on fuel displacement potential — Assess battery usage under V2G scenario— Emissions impacts of engine operation— Use travel data from five other municipalities

• Collect on-road data with PHEV research vehicle using several battery options and compare with simulation results

Acknowledgements

• Programmatic Support of FreedomCAR and Vehicle Technologies Program of the US DOE

— Tien Duong, Vehicle Technologies Team Lead— David Howell, Energy Storage Systems— Lee Slezak, Vehicle Systems

• Technical Support from East West Council of Governments (St. Louis, Missouri)

— Todd Barat