Two years of multi-scale simulation of laser welding for optimal battery pack manufacturing
In connection with two year anniversary of the European Union-funded project entitled “multi-scale simulation of laser welding for optimal battery pack manufacturing” with the heroic acronym “LaserBATMAN”, we found it appropriate to take a step back and reflect on the journey travelled and the road ahead.?
Project Background?
The aim of the project is to optimise the battery pack manufacturing process, targeting specifically the laser welding process. Simulation of different scales, ranging from the welding of battery terminals and busbars during manufacturing to the electrochemical performance of the assembled battery pack, will be performed to study and optimise the manufacturing process. The project consortium consists of University of Sk?vde (Sweden), Technical University of Denmark (Denmark), Volvo Group Trucks Operations (Sweden), Aurobay Powertrain Engineering Sweden (Sweden), and Resolvent (Denmark).?
?
Growing Battery Market for “BAT-mobiles”?
The battery demand is growing. Some of the expanding application areas are energy storage and electrical vehicles, EV’s (perhaps a more suitable acronym in this context, “BAT-mobiles”). The European Commission views batteries as a key technology for the transition to climate neutrality and to a more circular economy.??
Several of the consortium members, including resolvent, attended the Nordic Regional Battery Visit the 9th to 10th April 2024 in Mariestad/Sk?vde, Sweden. Here, the agenda was to strengthen the focus on battery manufacturing infrastructure and value chain within the Nordic region.??
One of the essential players, Volvo Group, being both a large industry and a consortium partner via the Trucks division, has recently initiated the process to establish a large-scale production plant for battery cells in Sweden. Optimisation of the battery welding process - as this project contributes to - is expected to be one out of many steps required in ensuring competitive battery manufacturing.?
The Year in Review?
In resolvent, we established a simulation framework that allows for optimisation of the design and operation of battery applications (battery packs) using Multiphysics simulation. Using scientific models, simulations allow prediction of temperature, electric potential, state of charge, and state of health of each battery cell after any cycle after historical cycle conditions (see Figure 2).
Apart from integration of advanced battery chemistry models and a knowledge-driven parameter fitting approach, numerical methods involving state-of-the-art machine learning techniques have been adopted to deliver high fidelity at low computational efforts. Figure 3 summarises the capabilities and techniques used in the overall simulation framework.?
?
The fundamental concepts of the simulation model are summarised in a blog on the resolvent webpage, and use case examples for optimisation were presented by Modelling specialist André Gugele Steckel during the COMSOL day on Optimisation.?
Stay tuned for more updates.??
Project Consortium Achievements?
Significant efforts have been put into establishing the empirical foundation of the project. Now, this foundation covers CT scan, visual inspections, and macroscopic examinations of experimentally performed welds for different welding process parameters.?
A high-fidelity model of the welding process has been established and fitted to the experimental results, thereby successfully predicting keyhole formations (see example, Figure 4). These results were published by project partners. This milestone serves as the basis for evaluation of welding process impact on battery pack lifetime performance.?
People For Success?
Looking back at many learnings and achievements after two-thirds into the project, we are confident in successfully achieving the milestones of the project, and we are looking forward to more of the great collaboration.?
Don’t hesitate to consider resolvent as a trusted simulation and engineering partner in your next project consortium.?
Acknowledgements?
The authors would like to acknowledge the financial support by the European M-ERA.NET 3 call (project9468 LaserBATMAN), Innovation Fund Denmark (grant number 1139-00001), and the Swedish Governmental Agency for Innovation Systems (Vinnova grant number 2022-01257).?