PosterPDF Available

Early Thermal Runaway detection in lithium ion batteries by using of a coupled electrical -thermal plausibility model

Early Thermal Runaway detection in lithium ion batteries by using of
a coupled electrical - thermal plausibility model
Jens Grabow*, Jacob Klink, Nury Orazov, Ralf Benger
Clausthal University of Technology, Research Center Energy Storage Technologies, D-38640 Goslar
[1] X. Feng et al., Energy Storage Materials (2018), doi: 10.1016/j.ensm.2017.05.013.
[2] Q. Wang et al., Progress in Energy and Combustion Science (2019), doi: 10.1016/j.pecs.2019.03.002.
Jens Grabow, M. Sc.
Am Stollen 19A, D-38640 Goslar
"ReserveBatt - Systemdienstleistungen für den sicheren
Betrieb des Energieversorgungssystems: Momentanreserve
mit Hochleistungsbatterien" FKZ: 03ET6123A
In addition to the high usable electrical energy, a large amount of chemical energy is stored in the cell
components, which can lead to safety-critical states in the case of incorrect handling or production defects [1]
As a result of exothermic side reactions and insufficient heat removal in the worst case, the self-accelerating
temperature increase can lead to thermal runaway, which can release toxic substances and large amounts of
heat [1,2]
This Safety-critical faults occur in practice (although scarcely), so that it is necessary to warn of an upcoming
thermal runaway as early as possible to minimize the risks for users and the environment
This work investigates the effect of the change in voltage caused by the temperature variation by using a
parallel running model as an indicator for an implausible cell state to detect a ongoing Thermal Runaway
About 255s after the cell has been heated
externally, the error is detected by a
temperature deviation of 4.5°C averaged
over 2 minutes
About 700s after the cell has been heated
externally, the fault is detected by a voltage
deviation of more than 0.065V over a period
of at least 4 minutes
Fault detection by temperature comparison
2545s (42 min) and by voltage comparison
2100s (35 min) before the first venting of
the Thermal Runway
1. Introduction
2. Method
3. Findings
3. Conclusion
The cell states temperature T and voltage U are estimated based on a coupled cell model (Model level) and compared against the
measured data (Reality). When these differences exceed a defined threshold the integrated Fault Detection will suspect a cell failure.
Both model validation and thermal fault induction are performed during highly dynamic cell load by the power demand of the WLTP
Fault detection
 
Early detection of a safety critical status through the plausibility comparison
of both voltage and temperature values
The temperature comparison shows an earlier and more robust response,
but requires temperature sensors
The voltage comparison is in principle more sensible to interference and
false detections, but it can theoretically detect various realistic faults and
does not require any additional sensors
The coupling of both signals is useful when developing a detection method
for practical applications
4. Future work
Transfer of this approach to module level
Verifying the sensitivity for other trigger
conditions, which lead to faster Thermal
Runaway conditions
Inspect the stability of the detection method
during long-term monitoring with changing cell
Identify and develop necessary model
t= 4057 s
t= 5138 s
t= 5237 s
dissipation Fault
 
 
 
Power demand
Electrochemical Model
Model level
Model coupling Thermal model
Electrochemical model
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.