sensor enhanced and model validated life extension of li ......dnv gl © safety prognostic approach...
TRANSCRIPT
DNV GL © SAFER, SMARTER, GREENER DNV GL ©
Sensor Enhanced and Model Validated Life Extension of Li-Ion Batteries for Energy Storage
1
DNV GL, NexTech Materials, Beckett Energy Systems
DNV GL ©
Value Propositions (derived from list of Battery Challenges)
(Safety) Prognostic approach for battery failure
Controlling battery packs to extract maximal capacity
Enables greater safety for fast charging
Assurance against “perceived” liabilities
General Safety and prevention – multiple inquiries
<5% system cost but > 5% value add?
Integration flexibility
Value in harsh climates
Cost vs. complexity
Non xEV applications (grid storage…)
Failure modes
Modularity of system
Battery and BMS Failure Modes
Expanding pack lifetime or operational limits
2
DNV GL ©
Safety Prognostic Approach to Battery Failure
3
240 245 250 255 260 265 2700
50100
150200250
300350400
Tem
pera
ture
(d
eg
C)
240 245 250 255 260 265 2700
2040
6080100
120140160
Sen
sor
Resp
onse
(kO
hm
)
Battery Surface Temp
Air Temp
240 245 250 255 260 265 2700
3
6
9
Volt
age
(V
)
Time (minutes)
240 245 250 255 260 265 2700
50
100
150
Sen
sor
Resp
onse
(kO
hm
)
Cell Voltage
Sensor Signal
Thermal Runaway Indicated
by Temperature Spike
Sensor Indication of Off-Gassing
Voltage
Indication of
Failure
Overcharge Test
DNV GL ©
Industry-Wide Problem
4 4
Automated Halogen
Extinguisher
DNV GL ©
Safety & Reliability
5
Safety and Reliability
Prognositics
Performance
Off gas
monitoring
DNV GL ©
Offgas Monitoring Enables Appropriate Use of Automated Fire Extinguishing
6
2 minutes
ahead of
voltage
7-8 minutes
total
7.4 minutes
ahead of
temperature
DNV GL ©
Failure Mode Identification
7
Voltage
2 Minutes
Tem
pera
ture
7 Minutes
Tota
l Early
Warn
ing
Regim
e
7 Minutes – Days
Safety Prognostics
Nail puncture
test – coincident
with V and T
DNV GL ©
Integration Flexibility, Cost vs. Complexity, and Modularity
1-3 sensors within a CES unit
8
• Pack level monitoring
• Benign offgas events are
detectable within minutes
• Catastrophic early warning
detectable < 1 min
DNV GL ©
Control Strategies to Convert to a Binary Signal
9
200 205 210 215 220 225 2300
500
Tem
pera
ture
(d
eg
C)
200 205 210 215 220 225 2300
200
Sen
sor
Resp
onse
(kO
hm
)
Battery Surface Temp
Air Temp
200 205 210 215 220 225 2300
5
10
15
20
Volt
age
(V
)
Time (minutes)
200 205 210 215 220 225 2300
100
200
Sen
sor
Resp
onse
(kO
hm
)
n=100, 2x std dev
n=100, 3x std dev
n=2000, 2x std dev
n=2000, 3x std dev
Voltage, 2x std dev
Voltage, 3x std dev
DNV GL ©
Detection Rates based on Diffusion of Offgas in CES Unit
10
DNV GL ©
Extraction of Maximal Capacity and Extension of Operational Limits
Use sensor as safety assurance for:
11
Increased energy services
Extending cell voltage limits
Expanding C-rate (power) capability
Extending Life or Cumulative Throughput
More value
from the
same system
More
Energy
More
Power
More
Throughput
DNV GL ©
What is the best path for cost reduction?
Monitoring and more aggressive
cycling (A)
Systems that have monitoring
and life extension (L)
Monitoring, life extension, and
more aggressive treatment (LA)
Life extension and downsizing
(LD)
Life extension and bypassable
modules (LB)
Life extension, downsizing, and
bypassable modules (LDB)
Life extension, aggressive
cycling, downsizing, and
bypassable modules (LDBA)
12
0
5
10
15
20
25
30
35
2010 2015 2020 2025 2030 2035 2040 2045
Lifetime Throughput
Monitored System Capacity (kWh)
ConventionalSystem Capacity (kWh)
Monitored 25th Capacity (kWh)
Monitored 75th Capacity (kWh)
Conventional 25th Capacity (kWh)
Conventional 75th Capacity (kWh)
$0.00$0.00
$0.05
$0.01
$0.01
$0.06
$0.00
$0.01
$0.02
$0.03
$0.04
$0.05
$0.06
$0.07
$0.08
Minimum $/kWh,Discharge Only
(Islanding)
Minimum $/kWh PaidCharge and Discharge
(Regulation)
Minimum $/kWh, PaidDischarge, Cost Charge
(Load Shifting)
$/k
Wh
CAPEX/Throughput
Sensor Max
Sensor Mean
Sensor Min
Conventional Max
Conventional Mean
Conventional Min
78%
-63%
-63%
-14%
Monitored System Life Output Improvement
"Islanding" Mean Cost Difference
"Regulation" Mean Cost Difference
"Load Shifting" Mean Cost Difference
Monitored System vs. Conventional System
$38,653
$49,500
Monitored System CAPEX Conventional System CAPEX
20
30
Monitored System Size(kWh)
Conventional System Size(kWh)
3,535,286
1,991,544
Monitored Discharge kWh Conventional Discharge kWh
Total Discharged kWh
DNV GL ©
Assurance against perceived and real liabilities
Perception of decreased safety in longer life
Real consideration for monitoring in enclosed spaces
Real consideration for monitoring in populated areas or the built environment
13
𝐴𝑣𝑜𝑖𝑑𝑒𝑑 𝐶𝑜𝑠𝑡 = 𝑝𝑑 ∗ 𝑝𝑐 ∗ 𝐶𝑐
𝐴𝑑𝑑𝑖𝑡𝑖𝑜𝑛𝑎𝑙 𝑅𝑒𝑣𝑒𝑛𝑢𝑒 = 𝐿 ∗ 𝑅𝑘𝑊ℎ ∗ 𝐸𝑑 ∗ 𝐿𝑒𝑥 (2)
Where pc is the probability of a catastrophe, pd is the probability of detection, and Cc is the costs of that catastrophe.
If the revenue potential per kWh is RkWh, and there is the ability to deliver energy at the rate of Ed (kWh/mo), the expected lifetime of a non-sensor equipped system is L (in months), and the potential life extension percentage is Lex,
Catastrophe
avoidance
Life Extension
DNV GL ©
Additional Cost Opportunity
14
𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝐶𝑜𝑠𝑡 𝑅𝑒𝑑𝑢𝑐𝑡𝑖𝑜𝑛
= 𝐵𝑎𝑠𝑒𝑙𝑖𝑛𝑒 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 − 𝐷𝑜𝑤𝑛𝑠𝑖𝑧𝑒𝑑 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 ∗$𝑐𝑎𝑝𝑎𝑐𝑖𝑡𝑦
𝑘𝑊ℎ− 𝑆𝑒𝑛𝑠𝑜𝑟 𝑆𝑦𝑠𝑡𝑒𝑚 𝐶𝑜𝑠𝑡
(4)
𝐶𝑜𝑜𝑙𝑖𝑛𝑔 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝐶𝑜𝑠𝑡 𝑅𝑒𝑑𝑢𝑐𝑡𝑖𝑜𝑛
= 𝐵𝑎𝑠𝑒𝑙𝑖𝑛𝑒 𝐶𝑜𝑜𝑙𝑖𝑛𝑔 − 𝐷𝑜𝑤𝑛𝑠𝑖𝑧𝑒𝑑 𝐶𝑜𝑜𝑙𝑖𝑛𝑔 ∗$𝑐𝑜𝑜𝑙𝑖𝑛𝑔
𝑘𝑊ℎ− 𝑆𝑒𝑛𝑠𝑜𝑟 𝑆𝑦𝑠𝑡𝑒𝑚 𝐶𝑜𝑠𝑡
(5)
Reducing
cooling load
Reducing
overcapacity
DNV GL ©
Non xEV Applications
15
Offshore oil & gas
- Dynamic Positioning Shuttle Tanker =18 MW for 30 minutes of backup
power for maneuvering during a blackout. Vessel is 300 m long.
- Drilling applications – 5-6 MW transients in <1 minute
- Harbor and offshore tugs: 7-8 MW electric only pulling
DNV GL ©
Harsh Environments < -- > Extended Performance
16
Higher temperature
Cooling load and burden
Physical, ingress, humidity, or mechanical hazards
DNV GL ©
Battery and BMS failure modes
17
Integrated with BMS Independent
of BMS
• Automated
Class D
Extinguisher
• Redundancy
• Maintenance
Alarms
• Distinguish
cell anomaly
from
catastrophic
failure
• Expanded
Operation
• “Pull back”
from
extremes
• Redundancy
in Shut Down
• Bypassable
Modules
DNV GL ©
SAFER, SMARTER, GREENER
www.dnvgl.com
18