Revolutionizing Battery Technology: AI and the Future of Energy Storage

The quality of batteries affects many aspects of our daily lives: Will we arrive at our holiday destination in our electric car? Can we be reached on our smartphones? Does the hearing aid still have plenty of power late at night? Mohamed El Bahnasawi is working on making batteries smarter and more efficient as part of the EU Horizon Europe project ‘Battery Cell Assembly Twin (BatCAT)’. The aim is to use artificial intelligence to learn more about the processes in the batteries in order to be able to use them more efficiently in the future.

Mohamed El Bahnasawi is not a chemist or physicist, but a researcher in the field of artificial intelligence. How are he and his colleagues approaching their research into more intelligent batteries in the BatCAT project? He explains this to us: “We create a digital twin of the battery. Using this twin, we can model everything: from the chemical process to the manufacturing. The digital model allows us to simulate multiple scenarios without having to work with real batteries in the physical world.”

Demand for smarter batteries is steadily increasing: “As batteries power a growing range of technologies, from smartphones to electric vehicles, improving their efficiency without increasing their size is essential. For example, smaller, more durable batteries could extend the range of electric vehicles or enable slimmer, more efficient smartphones. Progress in this field is rapid.” Artificial Intelligence provides the means to make predictions for the future based on historical data. The same should be possible for batteries, as a way of predicting their future behaviour and condition. To obtain the data, sensors and intelligent software must be implemented to help us understand the battery’s condition and the effects on its overall life cycle.

One important aspect of the project, which is funded by the European Union, is ‘explainable AI’. Mohamed El Bahnasawi elaborates: “Deep Learning is highly efficient. However, in many fields, events occur between data input and output that we cannot explain. We call this Black Box. The European Union is restrictive in this regard: We need to be able to describe and explain what the AI is doing. For us, this means finding a compromise between explainability and performance. We need a model that is explainable and also works very well.” Eighteen partners in nine European countries have been working on this since January 2024. The project is led by researchers at the Norwegian University of Life Sciences. At the University of Klagenfurt, Martin Gebser (Department of Artificial Intelligence & Cybersecurity) and Kyandoghere Kyamakya (Department of Smart Systems Technologies) are responsible for the project.

Mohamed El Bahnasawi is a member of the research staff working on the project and is also writing his doctoral thesis as part of his work. He completed his Bachelor’s degree in Mechatronics, Robotics and Automation Engineering at Egypt’s Nile University. A semester abroad took him to the University of Applied Sciences in Wels, Upper Austria. Eventually, he decided to enrol in the master’s degree programme in Information and Communications Engineering / Autonomous Systems and Robotics at the University of Klagenfurt with the support of the AAU-funded Technology Scholarship (organised with the OeAD). He then went on to combine the full-time degree programme with a full-time job: For one year, he worked as a hardware development engineer at AT&S in Leoben for his master’s thesis. He has been employed as a pre-doc researcher in the BatCAT project since June 2024.