MetalClass: MetalClass - AI-based real-time classification of metallic secondary raw materials using PGNAA.

Recycling the scrap available in Europe as a secondary raw material is the safest, most sustainable, most ecological and most economical form of raw material supply. Although metals can in principle be recycled an infinite number of times, remelting without loss of quality and increasing the secondary raw material content is only possible if the composition is recorded before remelting and interfering materials are removed from the recycling loop. Currently, the evaluation of scrap is done by samples that are considered representative.  If it were possible to record the elemental composition of scrap non-destructively in real time, input streams of the recycling process could be optimally controlled for the first time.

Due to the high and heterogeneous mass flows in copper and aluminum production, there is a great interest in the classification of recycling materials in real time in order to categorize them according to existing standards and regulations. According to the state of the art, there is currently no metrological solution for copper or aluminum production.

With this project, we are pursuing the goal of developing a measurement method for the non-destructive real-time classification of copper or aluminum scrap based on the PGNAA. The core of the innovation is the development of novel AI evaluation algorithms, which do not consider the measurement data reduced to individual peaks, but use the totality of the counting data.