ElringKlinger
AI weld-position detection for laser welding of EV cell contacting systems
ElringKlinger cut set-up of weld-position detection from days to hours with TRUMPF's image-based AI
Problem
ElringKlinger welds cell contacting systems for electric-vehicle batteries. The parts are up to 2 m long and carry more than 50 weld positions. Getting a reliable image-based detection of those positions used to take days.
Approach
A joining-technology engineer trains TRUMPF's EasyModel AI with VisionLine Detect on a few images. The images go to a cloud, the engineer marks the relevant areas, and the model finds component edges even at short cycle times. No programming is needed. The filter started at the Neuffen future-technologies plant and was rolled out to two more production sites.
Results
Takeaway
Training a vision model on a handful of images let a process engineer, not a programmer, set up detection for long parts with many weld points.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Vendor solution
- Evidence
- Vendor case study
Sources
Source summaries
TRUMPF's customer story describes how ElringKlinger uses EasyModel AI with VisionLine Detect to find weld positions on battery contacting systems. An engineer says good results now take hours, not days. The story names no start year and gives no cost figures.
Full details are in the original sources.