Energy Storage Monitoring System and In-Situ Impedance Measurement

ww

w.in

l.gov

Energy Storage Monitoring System and In-Situ Impedance Measurement

Modeling

Jon P. Christophersen, PhDPrincipal Investigator, Advanced Energy Storage Life and Health PrognosticsEnergy Storage & Transportation Systems

John L. Morrison, PhD, Montana Tech William H. Morrison, Qualtech Systems Inc.Chester G. Motloch, PhD, Motloch Consulting, Inc. Chinh D. Ho, Energy StorageHumberto E. Garcia, PhD, Monitoring and Decision Systems

May 15, 2012 Project ID ES122

2012 DOE Vehicle Technologies Program Annual Merit Review

This presentation does not contain any proprietary, confidential, or otherwise restricted information.

2

• Ongoing Project• Start – October 2007

• Cost• Performance• Abuse Tolerance and

Reliability• Life Estimation• Total project funding:

– DOE – 660k (through FY11)• Funding received in FY11:

– $450k• Funding for FY12:

– $400k

Timeline

Budget

Barriers

• Montana Tech of the University of Montana

• Qualtech Systems, Inc.

Partners

Overview

3

• Objective:– Develop advanced in-situ diagnostic and prognostic tools for more

accurate prediction of the state-of-health and remaining useful life of energy storage devices.

• Benefits:– Safety and reliability: Rapid identification of failure thresholds– Cost: Smarter replacement schedules– Life Estimation: Improved sensor technology for state estimation– Performance: Improved management systems based on battery

condition using both energy and power.

• Applications:– Automotive (EV, HEV, PHEV)– Military (field radio operations, warehouses, vehicles, etc.)– Other applications include NASA, electric utilities,

telecommunications, aeronautics, consumer electronics, etc.

Relevance

4

• Mission Areas:– Rapid, in-situ impedance spectrum measurement techniques using

hardware and control software – Impedance Measurement Box– Modeling and prognostic tools that smartly combine impedance

measurements with other observations to determine a more accurate definition of battery health.

– Hardware and software that directly interfaces with onboard battery technologies to smartly monitor and report health – Energy Storage Monitoring System.

• FY-12 Objectives:– Design and build a 50-V rapid impedance measurement system.– Improve calibration system of rapid impedance measurement.– Continue validation studies of rapid impedance measurement

techniques using commercially available lithium-ion cells.– Initiate incorporation of rapid impedance measurements into new

and existing modeling and prognostic tools.

Relevance (cont.)

5

• Impedance Measurement Box:– Develop rapid impedance spectra measurement techniques– Design low-cost hardware and control software– Implement measurements under no-load and load conditions

• Modeling:– Equivalent circuit modeling of rapid impedance measurements– Feature extraction (e.g., effective charge transfer resistance, etc.)– Estimate pulse resistance and power capability from impedance

measurements– Implement existing modeling capabilities (ES124) to develop

prognostic tools for arbitrary aging conditions

• Energy Storage Monitoring System:– Passive measurements (voltage, current, temperature)– Active measurements (rapid impedance spectra)– Incorporate models to estimate overall state-of-health (SOH) and

remaining useful life (RUL)

Approach

ModelingRapid Impedance

Measurement System (IMB)

Hardware Development

Software Development

Rapid Impedance Measurement Techniques

No Load / Load Applications

Feature Extraction / State Estimation

TLVT/BLE

DADT(ES096)

Energy Storage Monitoring

System (ESMS)

Acquire Active Measurements

Expert Learning, Diagnostics, and

Prognostic

Acquire Passive Measurements

SafetyABR(ES124)

SOH and RUL Etimation

Management and Control

Energy Storage Device

• Flow diagram of the overall ESMS concept:

6

Approach (cont.)


7

• Novel techniques have been developed to acquire wideband impedance within seconds.

– Harmonic Compensated Synchronous Detection (HCSD)• Low resolution, computationally intensive, short test duration

• Input signal is an octave harmonic sum-of-sines excitation current.

– The excitation signal has a durationof one period of the lowest frequency.

• Synchronously detect the magnitudeand phase of the captured voltage response time record.

• Long term validation testing was completed using Sanyo SAcells.

– Cells aged at 40 and 50°C with periodic HCSD measurements under both no-load and load conditions.

Technical Accomplishments and Progress

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

0 1 2 3 4 5 6 7 8 9 10

Curr

ent (

A)

Time (s)

SOS Signal

Fundamental Frequency (0.1 Hz)


8

• HCSDs under no-load conditions:– HCSD impedance spectra grow in both ohmic and effective charge

transfer resistance.– The rate of impedance growth is higher at 50°C, as expected.

Accomplishments and Progress (cont.)

0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040

0.028 0.033 0.038 0.043 0.048 0.053

-Imag

inar

y Im

peda

nce

(j oh

ms)

Real Impedance (ohms)

Char.RPT0RPT1RPT2RPT3RPT4RPT5

50°C - No Load0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040

0.028 0.033 0.038 0.043 0.048 0.053

-Imag

inar

y Im

peda

nce

(j oh

ms)


Char.RPT0RPT1RPT2RPT3RPT4RPT5

40°C - No Load


9

• HCSDs under no-load conditions (cont.):– The HCSD real impedance at the semicircle trough is highly

correlated with both HPPC discharge resistance and available power.

– r² ≥ 0.950 in all cases, most fits are r² ≥ 0.980 .


44

46

48

50

52

54

56

58

60

62

64

37 38 39 40 41 42 43 44 45 46 47 48

L-H

PPC

Pow

er a

t 500

Wh

(kW

)

HCSD Real Impedance (mohms)

40°C Cells (Under Load)

40°C Cells (No Load)



46

48

50

52

54

56

58

60

62

64

37 38 39 40 41 42 43 44 45 46 47 48

L-H

PPC

Dis

char

ge R

esis

tanc

e (m

ohm

s)

HCSD Real Impedance (mohms)






10

• HCSDs under no-load conditions (cont.):– HCSD results generally match standard EIS measurements in the

key mid-frequency range.


0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.030 0.040 0.050 0.060 0.070 0.080

Imag

inar

y Im

peda

nce

(-jZ

'')


Test Cell 4 (EIS)

Test Cell 4 (HCSD)

0.000

0.001

0.001

0.002

0.002

0.003

0.020 0.022 0.024 0.026 0.028 0.030 0.032 0.034

Imag

inar

y Im

peda

nce

(-jZ

'')


Test Cell 3 (EIS)

Text Cell 3 (HCSD)

Software Development

11

• HCSDs under no-load conditions (cont.):– The calibration issue has been resolved and the EIS and HCSD

measurements now match very well using test cell circuits.



12

• HCSDs under no-load conditions (cont.):– A new validation study is now underway with Sanyo SA cells to

measure HCSD at multiple DODs.– At RPT0, the mid-frequency semicircle generally increases in both

height and width with increasing DOD.


0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.025 0.030 0.035 0.040 0.045 0.050 0.055 0.060 0.065

Imag

inar

y Im

peda

nce

(-j o

hm)

Real Impedance (ohm)

Full Charge

10% DOD

20% DOD

30% DOD

40% DOD

50% DOD

60% DOD

70% DOD

80% DOD

90% DOD

Before Residual

Full Discharge

3.60

3.65

3.70

3.75

3.80

3.85

3.90

3.95

4.00

4.05

4.10

0 10 20 30 40 50 60 70 80 90

Volta

ge (V

)

Time in Profile (s)

Estimated Load Voltage


13

• HCSDs under load conditions :– Cells were cycled at elevated temperatures with HCSD

measurements triggered every 50 cycles.– Under-load measurements were affected by a non-constant DC

bias voltage.


-25

-15

-5

5

15

25

35

0 20 40 60 80 100

Pow

er (k

W)

Time in Profile (s)

PHEV Charge-Sustaining Profile

Engine-Off Pulse

LaunchPulse

RegenPulse

Cruise Pulse

Discharge

Charge

0.031

0.032

0.033

0.034

0.035

0.036

0.037

0.038

0.039

0.25 0.26 0.27 0.28 0.29 0.3 0.31

HCSD

Rea

l Im

peda

nce

at S

emic

ircle

Trou

gh (o

hms)

Cycle-Life Pulse Resistance (ohms)

Cell 13 (40°C)Cell 14 (40°C)Cell 15 (40°C)Cell 19 (50°C)Cell 20 (50°C)Cell 21 (50°C)

40°C Cells - 1.6 Hz50°C Cells - 3.2 HzEngine-Off Pulse

40 C: r²>0.9050 C: r²>0.97


14

• HCSDs under load conditions :– The Engine-Off impedance measurement under load still shows a

mid-frequency arc, but the Warburg tail has flattened out.– The growth in effective charge transfer resistance is linearly

correlated to the corresponding cycle-life pulse resistance.


-0.010

-0.008

-0.006

-0.004

-0.002

0.000

0.002

0.004

0.006

0.01 0.02 0.03 0.04 0.05 0.06

-Imag

inar

y Im

peda

nce

(j oh

ms)


250003000055000600008500090000115000120000145000150000

Engine-Off Pulse

0.1 Hz

1638.4 Hz

40°C

Time


15

• HCSDs under load conditions :– Despite the low-frequency effects, the height and width of the

effective charge transfer resistance semicircle is similar under both no load and load conditions.


0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040

0.028 0.033 0.038 0.043 0.048 0.053 0.058 0.063

-Imag

inar

y Im

peda

nce

(j oh

ms)


No Load (1 Period)

Under Load (1 Period)

Under Load (3 Periods)

0.1 Hz

3 kW

Hardware Development

16

• Hardware Development:– The 50-V Impedance Measurement Box, including prototype

hardware and upgraded control software, has now been completed.

– Upgraded features for this hardware system include:

• Applied dynamic voltage range between 0 and 50 V (i.e., system can test 5-V cells and 50-V modules)

• Significantly reduced noise crosstalk in voltage feedback loop when removing the DC bias (up to 50 V)

• Modular design for easier debugging and future upgrades• Voltage and current probe protection• Improved calibration technique


Feature Extraction / Training Regression

Battery Models- ABR (ES096)- DADT(ES124)- TLVT/BLE- Lumped-Parameter Models

Dynamic ParameterEstimation

Training Data

Dynamic ParameterPrediction

Online Measurements

Online ParameterInterpretation

Online Health Estimation/Prediction

SOH / RUL

Feature Extraction / Online Regression

• Modeling and State-of-Health Estimation:– Preliminary SOH assessment tool under development using results

from the HCSD validation studies.

Expert Learning and Prognostic Tools

17


18

• Montana Tech of the University of Montana:– Expertise in signal processing, hardware design– A professor and several graduate students have been involved in

this effort.

• Qualtech Systems, Inc.:– Expertise in software development

• Other (future) collaborators:– Agreement for a developmental collaboration between INL and the

Army (CERDEC) is being drafted.– Various discussions with private industry about IMB/ESMS

applications are also underway.

Collaborations

19

• Impedance Measurement Box:– Investigate possibility of lower frequency measurements while still

retaining fast measurement speeds.– Mitigate observed transient effects and bias voltage error for

under-load measurements.

• Modeling:– Develop dynamic parameter estimation and prediction tools for

online prognostics at arbitrary aging conditions.• Feature extraction from impedance spectra• New and existing modeling tools (e.g., ES124)

– Continue validation testing for enhanced modeling capability.

• Energy Storage Monitoring System:– Explore low-cost hardware solutions for embedded system

applications.– Seek collaborative opportunities with other groups involved in

battery prognostics and control.

Proposed Future Work

20

• The Impedance Measurement Box (IMB) enables low-cost, rapid, in-situ impedance spectra measurements.

– The IMB addresses cost, safety, performance, and life estimation barriers for energy storage devices.

• Significant IMB accomplishments:– The 50-V Gen 3 IMB hardware design has been completed.– Improved calibration techniques have been implemented for more

accurate impedance measurements.– Initial validation study is complete and a new study is underway

with rapid impedance measurements at multiple DODs.

• A preliminary state-of-health assessment architecture design has been developed.

– Dynamic parameter estimation and prediction rely extensively on battery modeling capability and training regression tools

Summary

TECHNICAL BACK-UP SLIDES


22

• Novel techniques have been developed to acquire wideband impedance within seconds:

– Impedance Noise Identification (INI)• Very high resolution, computationally very intensive, very long

test duration– Compensated Synchronous Detection (CSD)

• Lower resolution, computationally intensive, long test duration– Harmonic Compensated Synchronous Detection (HCSD)

• Low resolution, computationally intensive, short test duration– Fast Summation Transformation (FST)

• Low resolution, computationally simple, short test duration– Cross-Talk Compensation (CTC)

• High resolution, computationally intensive, short test duration

• INL, Montana Tech, and Qualtech Systems Inc. collaborated on the development of these techniques .

Impedance Measurement Box


23

• Purpose of the initial HCSD validation tests:– Demonstrate the effectiveness of the rapid impedance spectra

measurement technique under both no-load and load conditions using Sanyo SA cells.

• Cells aged at 40 and 50°C for 150,000 cycles with periodic reference performance tests (RPTs) to gauge degradation.

• HCSD measurements under no-load conditions conducted on all cells at each RPT.

• HCSD measurements under load conditions conducted on some cells during cycling.

– Designated cells were subjected to a total of 10,000 HCSD measurements under load with no obvious signs of additional degradation.

HCSD Validation Testing


24

• Chemistry:– Li(Mn, Co, Ni)O2 + Li-Mn-O spinel / Graphite

• Rated Capacity:– 1.2 Ah (C/1 Rate)

• Battery Size Factor:– 1400

• HPPC Voltage Range:– Vmax = 4.1 V– Vmin = 3.0 V

• USABC Application:– Minimum PHEV

Sanyo SA Cells

-0.002

-0.002

-0.001

-0.001

0.000

0.001

0.001

0.002

0.002

5 7 9 11 13 15

Volta

ge D

iffer

ence

(V)

Time (s)

Engine-Off Pulse


25

• The voltage bias profile for HCSD measurements under load can be observed with simulations.

• A bias voltage remains once the DC term is removed and it affects the measured spectra, especially at low frequencies.

Voltage Bias (Under Load)

3.60

3.65

3.70

3.75

3.80

3.85

3.90

3.95

4.00

4.05

4.10

0 10 20 30 40 50 60 70 80 90

Volta

ge (V

)

Time in Profile (s)

Estimated Load Voltage

Energy Storage Monitoring System and In-Situ Impedance Measurement

Documents

Transcript of Energy Storage Monitoring System and In-Situ Impedance Measurement