EUREKA SMART SIMS
| Reference number | |
| Coordinator | Luleå tekniska universitet - Luleå tekniska universitet Inst f teknikvetenskap & matematik |
| Funding from Vinnova | SEK 2 975 000 |
| Project duration | January 2023 - August 2026 |
| Status | Completed |
| Venture | Eureka Cluster within Sustainable Value Chains |
Important results from the project
The project goal was achieved, to create the prerequisites for a longer fatigue life of welded products. All subgoals were reached, generating advanced manufacturing data, better correlations between weld topology and fatigue life, new criteria, including for automated inspection, and demonstrating self-learning AI-based welding systems, all even transferable to AM. Given the ambitious goals, even clearer correlations are desirable due to the complexity of the mechanisms.
Expected long term effects
The developed knowledge and models of weld topology will reduce uncertainties related to fracture, extend product life, and enable lightweight designs. Self-learning manufacturing systems incorporating sensors, AI, and automated topology inspection will enhance productivity and lower risks. All of this contributes to increased competitiveness and a reduced environmental impact. The systematic framework on the complex mechanisms can be further developed to enhance knowledge and optimization.
Approach and implementation
Three manufacturing techniques were studied: arc welding, laser–arc hybrid welding, and Wire Arc AM. According to plan, seven companies and two research teams with complementary expertise initially tested various in situ and ex situ sensors. Systematic welding experiments and fatigue tests led to AI-assisted data analyses, theories of crack formation, and improved topology definitions, even including automated topology scanning. Self-learning manufacturing systems were demonstrated.