Electromobility BMW and the University of Zagreb Develop AI Models for Optimized Battery Cell Production

Related Vendors

dataTec AG

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

SIMTOS (Seoul International Machine Tool Show)

In the joint research project Insight, BMW and the Centre of Excellence for Robot Technology at the University of Zagreb are working on practical AI models to optimize battery cell production. The partners are developing along the entire value chain of battery cells: from electrode production to end-of-line tests and direct recycling developed in-house.

At the Munich Battery Cell Competence Center (BCCC), BMW develops battery cells for high-voltage batteries. This involves carrying out a series of tests that require a great deal of material and time. At the same time, these tests occupy the production facilities and laboratories. In other words, a great deal of material, space and costs are involved. This is where the Insight research project comes in: an artificial intelligence network uses previous test data and real-time data from ongoing production to precisely predict the process parameters and performance data of the battery cell. This allows the duration and number of test series to be significantly reduced while maintaining or improving quality. In individual process steps, the newly developed AI systems reduce the amount of material and time required by more than 50 percent.

Will the "Quarantine" Soon be Abolished?

The prediction models of the research project are not only used to reduce test series, but also to finally release battery cells. After the first charge at the end of production, the cells must be stored for a certain period of time at precisely defined temperatures before they can be installed in a battery housing. This phase is also known as "quarantine" and requires corresponding storage capacities. However, the AI systems in the research project are able to fully analyze the battery cells beforehand, meaning that this process step could be eliminated in the future.

AI Models Make Statements About Optimization Potential

Since the start of the project in 2024, the car manufacturer and the university have been developing joint solutions to improve the production of battery cells with the help of artificial intelligence. To this end, PhD students and students at the University of Zagreb are collecting and structuring existing production data. This data is used to create AI models that recognize certain patterns in the data. These AI models can then make predictions that further optimize production in terms of performance, quality and costs.

Profitable Partnership

The University of Zagreb is contributing its expertise in mechanical engineering, electrical engineering and computer science to this project. Both partners benefit from the continuous exchange of knowledge: the University of Zagreb offers the Munich-based company access to new research results, while the doctoral candidates and students benefit from the opportunity to apply their theoretical knowledge in practice.

Another aspect of this cooperation is the promotion of young talent. "Through the joint project, we are getting doctoral students and undergraduates interested in the fields of AI and battery cells as well as the exciting work in our battery cell competence centers," says Stefan Kerscher, Head of Battery Cell Technology Development at the BMW Group. "We are delighted when talented young people decide to start a career in our company." The students can use the cooperation to expand their professional network in the industry. (se)