Parametrization of Li-ion battery model via optimization

Background of thesis project


Physic-based battery models are important tool in understanding and development of battery technology as well as designing and improving control strategies. These models are reliable if only they are parametrized accurately. Unfortunately, not all applications allow an invasive parametrization via detailed experimental measurements, leaving us with an alternative method of parameter estimation via optimization. A well-structured optimization framework is proven to be capable of filling this gap, with an acceptable reliability, and it will facilitate building up more complex models.


Suitable background


This thesis is suitable for one engineering student with educational background within the field of Material Science, Electrical Engineering, Physics and/or Chemistry. The thesis works requires a good programming knowledge in Python. Understanding the physical, electrochemical, or control aspects of the Li-ion battery is a merit.

Description of thesis work
The scope of the work includes the following steps:
Review of the existing methods in the field (physic-based models plus optimization).
Create the initial full or reduced order model (optimization seed).
Building up the framework having DOE and parameters sensitivity in mind.
Optimization and exploring possible ways of validating the results.

Thesis Level:
Master of Science

Language:
English


Starting date:
January 2023


Number of students:
One


Tutor
Industrial tutor: Dr. Masood Tamadondar (masood.tamadondar@consultant.volvo.com), Gothenburg
Academic tutor: Prof. Göran Lindberg (gnli@kth.se), Royal Institute of Technology (KTH), Stockholm