Flanders Make
Flanders Make is the strategic research centre for the manufacturing industry. Our mission is to strengthen the long-term international competitiveness of the Flemish manufacturing industry. That’s why we work together with SMEs and large companies on pre-competitive, industry-driven technological research, resulting in concrete product and production innovation in the vehicle industry, the manufacturing industry, and production environments.
Problem Context
The value of an electric vehicle is currently largely determined by the cost of its battery. Hence it is important, especially for fleet operators, to determine the current state of health of the batteries in their vehicles and to understand how long these batteries will last and if any action can be taken to prolong the useful life of the batteries. Further, the cost-effective second life application relies on fast, low-cost, and reliable diagnostic methodologies makes it possible to reduce the remanufacturing time of second-life batteries. However, battery ageing is a challenging topic which is as of yet not fully understood by the battery community. Currently, there are several different approaches to modelling and predicting ageing, where the most popular methods are either purely data-driven (machine learning and AI) or physics based.
Goal of the thesis
The most important goal is that the explored approach provides better performance or novel capability compared to existing methods. In particular, the goal of this thesis project is twofold: 1) explore how the data-driven methods can be improved through physics informed features or how the strengths of data-driven and physics-based methods can be leveraged and combined to overcome their individual shortcomings and together produce more accurate predictions and 2) how the innovative method (such as electrochemical impedance spectroscopy based health estimation) correlates to existing capacity state estimation based methods.
You will be predicting battery ageing through a combination of data-driven and physics-based methods. Both methods have potential, but also come with challenges. To reach their full potential, data-driven methods need high quality data in high quantity. However, this data is challenging to obtain. Lab data is often of high quality, but low in quantity and expensive to produce (especially if one wants to cover all potential conditions (e.g., depth of discharge, temperature, humidities, etc.). On the other hand, field data is rich in its diversity and quantity, but lacks in quality, as gaps in the data, rogue datapoints and flawed measurements create challenges. Physics-based methods, on the other hand, face the challenge of requiring very specific data which can only be obtained through cell teardown. The challenge is to combine the best of these two methods.
Technical Approach
Specifically, the following tasks will underpin your master’s thesis:
* Literature survey on the status of data-driven and physics-based ageing prediction
* Select an approach and design a predictor
* Implement in Python and run on Flanders Make’s existing testing data.
* Conduct extra testing to collect battery cell characteristics when needed (equipment available at Flanders Make).
* Perform analyses and draw conclusions on the performance of innovative methods against existing methods. These can include, for instance, electrochemical impedance spectroscopy based health monitoring against battery capacity estimation based health monitoring, or one step further, degradation behavior like Loss of Active material (LAM) or Loss of Li inventory (LLI).
Profile Student
* Bachelor’s degree in a relevant field like Mechanical, Electrical Engineering, Physics, Chemistry or Data Science;
* Currently pursuing a master's in a relevant field like energy/electrical engineering, data science, AI, electrical engineering, statistics;
* Knowledge of batteries (e.g., electrical engineering and/or chemistry) and machine learning is preferred;
* Experience with Python, Matlab/Simulink, and machine learning is preferred;
* Passionate about research and new technologies with a focus on applications for energy storage systems;
* Passionate about research and new technologies with a focus on applications for machines or mechatronic systems of the companies.
This assignment can only be executed by a thesis student from a Belgian university.
Practical Data
* Thesis: This assignment is a topic for a master thesis for a Belgian university.
* Follow up moments take place at the offices of Flanders Make located in Leuven, Belgium.
#J-18808-Ljbffr