AlloySort: Real-time PGNAA analysis of metallic alloys for downstream targeted sorting

With regard to environmental protection, sustainability, economy and political requirements, recycling processes are becoming increasingly important and aim to achieve a significantly higher use of secondary raw materials.
AlloySort builds on the MetallClass project, which mainly dealt with the classification of different metal alloys.

There is currently no method in the copper and aluminum industry for non-destructive, real-time analysis of heterogeneous recycling materials that can also determine the exact composition of mixed scrap. Using Prompt Gamma Neutron Activation Analysis (PGNAA), we want to overcome this challenge, even with complex material flows or material batches.

We aim to integrate the system into the production process in the copper and aluminum industry by coupling it with sorting and conveyor technology. For this purpose, a conveyor belt is integrated into a demonstrator measuring system. The core of our approach is the development of new AI-based evaluation methods to determine the material compositions from the high-resolution, but very noisy, sensor data and to use these for sorting on the conveyor belt.