Numerical Simulation and Experimental Validation for Thermal Runaway on Lithium-ion Cells

Loading.... (view fulltext now)

Volltext

(1)

Wissen für Morgen

Knowledge for Tomorrow

Numerical Simulation and Experimental Validation

for Thermal Runaway on Lithium-ion Cells

N. Tanaka*

,1

, J. Mehne

2

, W. Nowak

2

, M. A. Danzer

3

, H. Döring

3

, W. G. Bessler

4

1Institute of Technical Thermodynamics, German Aerospace Center (DLR), Stuttgart, Germany

2Institute for Modeling Hydraulic and Environmental Systems, University of Stuttgart, Stuttgart, Germany 3Zentrum für Sonnenenergie- und Wasserstoff-Forschung Baden-Württemberg (ZSW), Ulm, Germany

4Offenburg University of Applied Sciences, Offenburg, Germany *nanako.tanaka@dlr.de

Motivation

Approach

Micro Model

Stochastic Model

Macro Model

Experiment

2013: Boeing

dreamliner battery

2006: DELL recalls

4 mio. laptop batteries

Heat

Thermal

Runaway

Triggering event Manufacturing (e.g. particles) Dendrite formation Internal short-circuit Thermal runaway mechanism Crash Over- charge Over- discharge External short-circuit Over- heating

Runaway = Chemistry + Heat transport

Li+ e– Microscale Macroscale 3D thermal model Electrochemistry & heating model

T QT Electro- chemical and micro-structural parameters Thermal and macro-structural parameters Trigger and runaway simulation Monte Carlo Stochastic para-meter variation Scale interface Experiment s Trigger of extreme events Model validation Model validation

Goal: Early-alert risk-aware battery management system

Stochastic and optimal model-predictive control with constraints

Complete sequential procedure (simplified):

Update of uncertain model predictions with measurements via Bayes‘ theorem:

System model:

Measurement model:

Solution of model equations with a particle filter:

• Continuous probability density is discretized by particles (individual model runs)

• Measurement update via reweighting of the particles

prediction information loss information gain update Bayesian filtering Monte Carlo prediction of surface temperature over time

𝑝 𝑥𝑡|𝑦0, ⋯ , 𝑦𝑡 = 𝑝 𝑦0, ⋯ , 𝑦𝑡|𝑥𝑡 𝑝 𝑦0, ⋯ , 𝑦𝑡 𝑥𝑡 = 𝑓 𝑥𝑡−1, 𝜇𝑡 𝑦𝑡 = 𝑔 𝑥𝑡, 𝜈𝑡

𝑥𝑡 ⋯ model state at time t 𝑦𝑡 ⋯ measurement at time t 𝜇𝑡 ⋯ model error 𝜈𝑡 ⋯ measurement error Simulation of self-accelerating heat with SEI decomposition and formation reaction at adiabatic condition. The initial temperature is 400 K. Simulation of differential scanning calorimetry (DSC) for

SEI decomposition and formation. Dash line : experiments by

Pasquier et al.* , Solid line s: simulation.

Heat rate : 5K/min.

1D simulation

3D simulation 3D simulation is compared with 1D simulation under nominal

discharge operation in 1 hour (1C rate).

3D, 2D and 1D model of single cell will be investigated using COMSOL

Temperature distribution of 3D cell model

General characterization

Battery cycling Abuse experiments such as

short circuit, nail

penetration and overcharge will be conducted.

Operation

characteristics of SONY US26650VT

Degradation models at high temperature include:

• Solid electrolyte interface (SEI) decomposition (CH2OCO2Li)2  Li2CO3 + C2H4 + CO2 + 0.5 O2

• SEI formation (Electrolyte decomposition)

2 C3H4O3 (EC) + 2 e– + 2 Li+  (CH2OCO2Li)2 + C2H4 • Electrolyte evaporation

C3H4O3 (liquid)  C3H4O3 (gas)

* J. Electrochem. Soc., Vol. 145, No. 2, 1998

Exothermic reaction

Abbildung

Updating...

Referenzen

Updating...