High Throughput Numerical Simulations with HPC and AI for Precision Engineering in Welding and Additive Manufacturing

Short description of the business experiment

The SHAPE-AM initiative aims to revolutionise welding and metal additive manufacturing processes by conducting high-throughput finite element simulations and collecting results to train a physics-informed artificial intelligence model. The project addresses the challenge of optimising these processes by developing a unique workflow that integrates experimental data from robots and sensors with advanced HPC simulations to perform real-time thermomechanical analyses.

By leveraging the RobTrack system and sophisticated AI models, SHAPE-AM aims to predict metal behaviour and provide insights for various processing conditions. This innovative approach will accelerate time-to-market, reduce process development costs, and minimise environmental impact while enhancing manufacturing efficiency and product quality.

The ultimate goal is to establish a new industry standard by combining HPC capabilities with advanced manufacturing technologies.

Organisations involved: 

End User: 3D-Components AS (3DC)

HPC provider: Sigma2

Domain expert: SimEvolutiom A/S 

Sigma2 is a member of the NCC Norway.