Multiscale Models and Model Reduction Methods for Predicting the Lifetime of Lithium-Ion Batteries

Aim of the BMBF Project: MULTIBAT

The development of efficient mathematical methods for the modeling of multi-phase and multi-scale processes in lithium-ion batteries.

 

The Modeling Involves:

  • the stochastic structure modeling of porous electrodes for the generation of parametrized dynamic structure models based on imaging data,
  • the mathematical modeling of the underlying physical and electrochemical processes by coupled systems of partial differential equations,
  • their efficient high-resolution simulation with modern adaptive multiscale methods,
  • as well as the model order reduction of high-dimensional, parameterized multiscale systems to enable rapid parameter studies, sensitivity analyses or uncertainty quantification.

The work focuses on the adaptive modeling of degradation processes that lead to a change in the underlying geometric structure of the porous electrode material. In particular, the local deposition of metallic lithium (lithium plating) is investigated in this project.

Main Task: Simulation Method for Degradation Prediction

Within the scope of the project consortium, it was our task to integrate the different-scale and different-dimensional mathematical algorithms and models developed within the project into a mathematical simulation method for the degradation prediction of Li-ion batteries.

In this case, for example, the interfaces already existing at the WWU for traditional RB methods are linked and extensions or new interfaces are created. In addition to the prototypical interfaces of the BEST software and the CoRheoS platform, the models developed for the multiscale numerics and model reduction are made available on the basis of the prototypical interfaces available at the ITWM.

Project Management

Prof. Dr. Mario Ohlberger, University of Münster

Industry Partner

Dr. Andres Pott, Deutsche ACCUmotive GmbH & Co. KG

 

Webside Project MULTIBAT