Master Thesis: Developing an online estimator for battery temperature

Within the Electromobility department, we work with the design and development of electromobility systems and components such as energy storage, motor drives, and charging solutions for Volvo group products. Electromobility is an important technology area with a strong contribution to reducing the CO2 emissions in our society.
The monitoring of Li-ion battery temperatures is essential to ensure high efficiency and safety. For Li-ion batteries, the desirable working temperature range (nominal performance) is [25 ͦC – 35 ͦC]. Nevertheless, the actual working temperatures of the battery cells deviate substantially from these nominal values, causing significant impairments in the battery performance. Battery lifespan decreases at high temperatures and its capacity drops at low temperatures. Furthermore, an uncontrolled increase of temperature in the cells causes thermal runaway with hence serious hazards.
Since the relationship between battery temperature and battery cell properties is nonlinear due to complex electro-chemical processes, machine learning methods such as Artificial Neural Networks (ANNs) can be used to create a black box correlating the battery temperature and a set of defined features such as transient coolant temperature, current profile, and environment conditions.


Suitable background


Master student with a background in System Controls & Mechatronics, Applied Mechanics, or any other relevant engineering program. You have a good background in Data analytics and Machine Learning as well as a basic knowledge in heat and mass transfer is counted as an extra merit.


Qualifications:
MSc Engineering student in System Controls & Mechatronics, Applied Mechanics, or any other relevant engineering program. Ready for thesis work 2023.
Experience in using Simulink for modelling, GT-SUITE, and simulation
Knowledge about machine learning methodology
Basic understanding of thermodynamic concepts such as heat transfer
Some scripting experience in e.g. Python, MATLAB
Communication skills and ability to document a work process in steps


Description of thesis work


The master thesis project aims at developing a data-driven black-box model that can give rapid and accurate estimation of a battery temperature. The thesis consists of multiple work packages, starting by literature study, define features and running a set of simulations in GT-Suite environment to create a dataset needed for model training and validation. The model will later be calibrated with experimental data at Volvo Trucks. The final goal is to provide a more reliable tool to anticipate and predict the battery temperature.




Thesis Level: Master

Language: English


Starting date: January 2023


Number of students: 1


Tutor
Masih Khoshab, masih.khoshab@volvo.com
Amirreza Movaghar, amirreza.movaghar.2@volvo.com


Academic Supervisor:
Torsten Wik, System och Reglerteknik at Chalmers, tw@chalmers.se
https://youtu.be/aUNiVxmiJDE