Next‑day delivery has become standard, and logistics fleets are shifting to electric trucks to cut fuel costs and emissions. The bottleneck is battery pack lifetime on long highway routes. A European consortium tackled this with multiscale simulation and a custom app built in COMSOL Multiphysics.
From laser weld physics to pack‑level design
Engineering consultancy Resolvent started by modelling how laser welding of busbars affects cell electrochemistry and degradation. What began as a weld‑focused multiphysics study evolved into a full cell‑to‑pack performance model that doubles as a design tool.
Their workflow:
- Build an electrochemical cell model, entering reaction‑kinetics equations
- Use the cell‑layer simulation results to train a deep neural network (DNN)
- Deploy this reduced‑order model inside a pack‑level simulation and wrap it in a custom simulation app
The DNN‑based model runs fast enough to explore many design variants while still capturing key cell‑level effects.
“Tool within a tool” for busbar welds
In the app, engineers can set pack geometry, number of cells, ambient temperature, and charge/discharge current profiles. A dedicated Busbar Weld Design Tool lets users sweep weld shape, busbar thickness, and laser penetration depth.
With one click, the app evaluates how these parameters impact:
- Weld temperature distribution
- Pack state of charge (SoC)
- SEI layer thickness and degradation indicators
Engineers can also add cooling airflow and immediately see turbulence, stagnant zones, and their effect on weld temperature and ageing. If improved cooling yields the same performance with fewer cells, the app quantifies that design trade‑off.
By encapsulating complex multiphysics models in an intuitive UI, the LaserBATMAN app gives OEM R&D teams a fast way to design more robust, long‑range truck battery packs and support wider adoption of green transport.
Source: IEEE Spectrum
