Online Estimation of Lithium Ion SOC MultiscaleFiltering ...€¦ · 10/27/2011 SHRM Lab., Seoul...
-
Upload
phungkhuong -
Category
Documents
-
view
212 -
download
0
Transcript of Online Estimation of Lithium Ion SOC MultiscaleFiltering ...€¦ · 10/27/2011 SHRM Lab., Seoul...
Sep 26, 2011
Online Estimation of Lithium‐Ion Battery SOC and Capacity with
Multiscale Filtering Technique for EVs/HEVs
BATTERY MANAGEMENT SYSTEMS WORKSHOP
Chao Hu1, Byeng D. Youn2, Jaesik Chung3 and T.J. Kim2
1 Department of Mechanical Engineering, University of Maryland at College Park2 School of Mechanical and Aerospace Engineering, Seoul National University
3 PCTEST Engineering Laboratory
Background and MotivationBackground and Motivation1
Task 1: Battery Health Diagnostics Task 1: Battery Health Diagnostics 2
Task 2: Battery Health PrognosticsTask 2: Battery Health Prognostics3
Task 3: Battery Health ManagementTask 3: Battery Health Management4
Concluding RemarksConcluding Remarks5
Contents
10/27/2011 2SHRM Lab., Seoul National University
Background and Motivation
10/27/2011 3SHRM Lab., Seoul National University
Battery Management System (BMS)
2. Battery Monitoring Module
SOC, SOH, SOL Module
Decision Logic
SensoryData
V, I, T
Control Signals
Charging & Equalization
Power In
External Charger or Alternator
4. Thermal Management
System
Battery
Engine
HCU
Generator
Battery Pack
Cells
…3. Battery
Control Module5. Safety Module
1. Module Sensing System (V, I, T measurements)
2. Battery Monitoring Module
3. Battery Control Module
4. Thermal Management System (cooling pumps/fans)
5. Battery Warning Device and Safety Module
1. Sen
sing
System
Hubwheel
Background and Motivation
10/27/2011 4SHRM Lab., Seoul National University
Battery Health Management (BHM) for BMS
Schedule‐BasedBattery Replacement
BHM‐Based
BHM‐Based
RUL1 > RUL2
Late Replacement (Failure)
Timely Replacement (Safe)
Average Co
st
Failu
re Co
st
Health Con
ditio
n
Lifetime
Case1
Case 2
+
–
Passive capacity Adaptive capacity (BHM)
+
–
Benefit 1. Cell balancing in multi‐cell battery chains
Benefit 2. Cell health management
Case 1: Enabling use of full cell capacities
Case 2: Anticipating & preventing future failures
Maximizing charging & discharging pack capacity
20%
Cdis
10%
60%
Cchar
40%
u44
h1
Slide 4
u44 It is not clear how short-term and long-term benefits can be distinguished.user, 9/21/2011
h1 The titles have been revised to be more specific. huchaostu, 9/22/2011
Overview of Battery Health Management
10/27/2011 5SHRM Lab., Seoul National University
Battery Cycle Testing
SOC & Capacity Estimation
Task I
Battery Health Diagnostics
Battery Cells Cycle Tester
Test con
ditio
n
SOC : Micro EKF/UKF
Capacity : Macro EKF/UKF
Volta
ge
Curren
t
SOC
Capacity
SOC
Projectio
n
t
End time
Task II
Battery Health Prognostics
Remaining Useful Time
Remaining Useful Life
Voltage Projection
Capacity
Operation History
RUT Estimate
t
Volta
ge
3.0V
4.2V
Current time
Prediction
UDDS
Battery Health Database (Offline)Ca
pacity
Cycle End life
Temperature
Discharge rate
Chemistry
Background health knowledge
Predicted RUL
Particle Filter Projection (Online)
Task III
Battery Health Management
Cell Balancing
Cell Replacement
+
–
Unbalance: Cdis = 20%, Cchar = 10%
+
–
Balance: Cdis = 60%, Cchar = 40%
Boosting/shuffling
Prob
ability
…Designed life
t
Cell 1 life
Cell Nlife
Cell to replace
Overview of Battery Health Management
10/27/2011 6SHRM Lab., Seoul National University
Battery Cycle Testing
SOC & Capacity Estimation
Task I
Battery Health Diagnostics
Battery Cells Cycle Tester
Test con
ditio
n
SOC : Micro EKF/UKF
Capacity : Macro EKF/UKF
Volta
ge
Curren
t
SOC
Capacity
SOC
Projectio
n
t
End time
Task II
Battery Health Prognostics
Remaining Useful Time
Remaining Useful Life
Voltage Projection
Capacity
Operation History
RUT Estimate
t
Volta
ge
3.0V
4.2V
Current time
Prediction
UDDS
Battery Health Database (Offline)Ca
pacity
Cycle End life
Temperature
Discharge rate
Chemistry
Background health knowledge
Predicted RUL
Particle Filter Projection (Online)
Task III
Battery Health Management
Cell Balancing
Cell Replacement
+
–
Unbalance: Cdis = 20%, Cchar = 10%
+
–
Balance: Cdis = 60%, Cchar = 40%
Boosting/shuffling
Prob
ability
…Designed life
t
Cell 1 life
Cell Nlife
Cell to replace
Concluding Remarks
10/27/2011 7SHRM Lab., Seoul National University
Multiscale Extended Kalman Filter (EKF) Proposed
Two Unique Strategies in Capacity Estimation
Multiscale estimation of SOC and capacity with time‐scale separation
State projection scheme for accurate and stable capacity estimation
Improved Performance
Enhanced accuracy (less noise and more stable) in capacity estimation
Higher efficiency by reduced frequency (1/L) in capacity estimation
Future Research
Extension of the proposed algorithm from a cell level to a pack level
Validation of the proposed algorithm with more extensive accelerated life test
u50
h3
Slide 7
u50 Highlight the uniquenesses of our proposed method: multiscale, stochastic, etc.
Express the uniquenesses, contributions, advantages of the proposed method.user, 9/21/2011
h3 The conclusion has been revised to reflect the uniquenesses, contributions and benefits of the proposed multiscale filtering method. huchaostu, 9/22/2011
Acknowledgements
10/27/2011 8SHRM Lab., Seoul National University
• PCTEST Laboratory (Battety Certification Firm, Columbia, MD)
• National Research Foundation (NRF), Korea (2011‐2014)
• Hyundai Motors R&D, Electric Vehicle Department, Korea (2011‐2012)
u59
h8
Slide 8
u59 Highlight the uniquenesses of our proposed method: multiscale, stochastic, etc.
Express the uniquenesses, contributions, advantages of the proposed method.user, 9/21/2011
h8 The conclusion has been revised to reflect the uniquenesses, contributions and benefits of the proposed multiscale filtering method. huchaostu, 9/22/2011
10/27/2011 10SHRM Lab., Seoul National University
Task 1: Battery Health Diagnostics
Time‐Scale Separation in State of Charge (SOC) and Capacity EvolutionsCa
pacity (A
h)
Time (month)
SOC (%)Capacity (Ah)
3 5 7 9 111
SOC (%
)
Time (hour)2 3 4 5 61
Capacity: slowly time‐varying health condition (macro time‐scale);
SOC: fast time‐varying system state (micro time‐scale)
Multiscale EKF – Hybrid of Coulomb counting and adaptive filtering technique
10/27/2011 11
Contribution 1 – Multiscale EKF for SOC & Capacity Estimation
Time updatexk,l–=xk,l‐1^ +ηi∙T∙ik,l‐1/Ck‐1
Ck–
xk,l–xk,l‐1^
Ck‐1
ik,l‐1 Measurement update(cell dynamic model)
xk,l^
Vk,l ik,l
Ckxk,L~
Macroscale achieved (go to Macro EKF)? xk,l^
No
Yesxk,L^(SOC)
Micro EKF
Macro EKF(Capacity)
SHRM Lab., Seoul National University
Cell current profile Cell voltage profile
Zoom of first minutes
SOC estimation Capacity estimation
Cell current
Zoom of first minutes
SOC estimationCell voltage
Task 1: Battery Health Diagnostics
Measurement updateCk+ = Ck– +KkC∙[xk,L^ – xk,L~ ]
Capacity transitionCk– = Ck–1+
SOC projectionxk,L~ = xk,0+η∙T∙ik,0:L‐1/Ck–
10/27/2011 12SHRM Lab., Seoul National University
Contribution 2 – State Projection for Accurate Capacity Estimation
t
SOC/x
tk,0 tk,L
Ck‒(L)Ck‒(N)
Ck‒(S)1
Microscale: T
Macroscale: L∙T
2
Ck−1(L)
Ck−1(N)Ck−1(S)
3
Ck‒(L)
Ck‒(N)
Ck‒(S)
1 Capacity transition
2 SOC projection (coulomb counting)
(SOC)Micro EKF
Macro EKF(Capacity)31 2
Task 1: Battery Health Diagnostics
3 Measurement update
Estimated SOC
10/27/2011 13SHRM Lab., Seoul National University
• Testing cells: Li‐ion prismatic cells (around 1.50Ah)
• Testing facilities: Arbin BT2000 cell tester with Espec SH‐241 temperature chamber at 25oC
• Test step 1: Setup UDDS test System• Test step 2: Program and load UDDS test profile into cell tester• Test step 3: Start test and acquire test data (duration: around 20hrs)
Arbin Cell Tester
Experiment Setup – UDDS Test System
Data Acquisition Device Temperature Chamber
Prismatic Cells Prismatic Cells
Test Jig
Task 1: Battery Health Diagnostics
Test Steup – Urban Dynamometer Drive Schedule (UDDS) Test
10/27/2011 14SHRM Lab., Seoul National University
Task 1: Battery Health Diagnostics
Test Result – Urban Dynamometer Drive Schedule (UDDS) Test
Compu
tatio
nal
time (s)
MEKF
36.2% saving
22 Comparison of efficiency (overall)
DEKF
5.813
3.711
MATLAB Version 7.11.0.584 on Intel Core i5 760 CPU 2.8 GHz and 4 Gbyte RAM.
11 Comparison of accuracy (small initial)
RMS error (mAh
)
MEKF
41.7% error reduction
DEKF
108
63
Average RMS errors after convergence (at t = 200min)
Capa
city Estim
ation
(Five Ce
lls)
MEKF
DEKF
Small initial (convergence check )
MEKF
DEKF
Accurate initial (stability check )
u53
h4
Slide 14
u53 We need to hightlight the benefits and advantages of our approach far more clearly.user, 9/21/2011
h4 1. Capacity estimation results of MEKF and DEKF are plotted in the same figure to deliver a more clear comparision.
2. Quantitative comparison results have been added to clearly show the superior performance of MEKF. huchaostu, 9/22/2011
0 100 200 3001.2
1.3
1.4
1.5
1.6
Cycle No.
Cel
l Cap
acity
(Ah)
Cell 1Cell 2Cell 3Cell 4
0 100 200 3001.2
1.3
1.4
1.5
1.6
Cycle No.
Cel
l Cap
acity
(Ah)
Cell 5Cell 6Cell 7Cell 8
0 100 200 3001.2
1.3
1.4
1.5
1.6
Cycle No.
Cel
l Cap
acity
(Ah)
Cell 9Cell 10Cell 11Cell 12
0 100 200 3001.2
1.3
1.4
1.5
1.6
Cycle No.C
ell C
apac
ity (A
h)
Cell 13Cell 14Cell 15Cell 16
Preliminary Results – Accelerated Cycle Life Test (Ongoing)
Task 2: Battery Health Prognostics
10/27/2011 15SHRM Lab., Seoul National University
1.0C Charging,1.0C Discharging
1.5C Charging,1.0C Discharging
1.0C Charging,2.0C Discharging
1.5C Charging,2.0C Discharging
Charging Rate
DischargingRate
Number of Cells
1.0C 1.0C 4
1.5C 1.0C 4
1.0C 2.0C 4
1.5C 2.0C 4
C = 1.5A; Temperature = 25oCOne cell to check 1.5C charge and 2.0C discharge
10 charging & discharging cycles
… …
Capacity and impedance check
u45 u46
h6
Slide 15
u45 It is not clear what we can observe out of the results.Also not clear correlation between C-rate and degradation.Please analyze these results to get more insightful understanding.user, 9/21/2011
u46 Can we update the results with recent cycle data?This could be critical in order to make some conclusions.user, 9/21/2011
h6 The results have been updated with recent cycle data. At this stage, it is still very difficult to see a clear correlation between the C-rateand capacity degradation, since cells (9-16) with higher C-rates exhibit similar degradation behavior compared to cells (1-8) with lowerC-rates. This slide can be used for the purpose of demonstrating our ongoing cycle life testing with preliminary results.huchaostu, 9/22/2011
Preliminary Results – Hybrid Algorithm for SOL Prediction (Ongoing)
Task 2: Battery Health Prognostics
10/27/2011 16SHRM Lab., Seoul National University
SOC&Capacity
Remaining Life
Health Diagnostics
Health Prognostics
Cell Dynamic Model Training
Degradation Model Training
Offline Process
Health Filtering(Mutiscale EKF)
Health Projection (Particle Filter)
Online Process
100 250200150
350
300
250
200
150
Time (cycles)
10% error bounds
300 350
100
50
RUL (cycles)
Life Prediction
Performan
ce Plot
0 100 400300200
1.6
1.5
1.4
1.3
1.2
Time (cycles)
1.5C Charging, 1.0C Discharging
500 600
1.1
1.0
Prediction at 300cycles
Test data
Capacity (A
h)
True RUL
RUL PDF
4000
RUL prediction(mean)
u58
h7
Slide 16
u58 Not clear how this result comes from.Is it based on someone else's results or ours?The workshop has many participants in this research. It is not a good idea to present others' results.user, 9/21/2011
h7 The life prediction approach with Particle Filter has been developed by researchers from NASA. But this integrated framework combining health diagnostics (with MEKF) and health prognostics has not been attempted. Thus, we may present this integrated framwork and show some preliminary results (shown in the bottom figures) to support it. huchaostu, 9/23/2011
Module 1 (5 Cells)
Cell‐Module‐Pack Health Information Panel (Conceptually Created)
Task 3: Battery Health Management
10/27/2011 17SHRM Lab., Seoul National University
0
100
1 2 3 4 5
0
10
State of charge (%)
Remaining life (yr)
Cell ID
Module 2 (5 Cells)
0
100
6 7 8 9 10
0
10
State of charge (%)
Remaining life (yr)
Cell ID
Pack (2 Modules)
Remaining useful time hours min.4 30 Cells requesting
replacement 5,7Real‐time & user‐friendly health information feedback for effective cell balancing/replacement
u49
h5