Sensor Controlled Quality Control and Sorting of Scrap Aluminium Alloys

Sensor controlled sorting is a promising technique for the upgrading of scrap aluminium alloys. In a universal detection system of Delft University characteristic features are extracted · from the material particles by means of cameras. Examples are size, texture, edge-structure, homogeneity, and colour pattern. Additional sensors are added for higher selectivity: electromagnetic detection, light transmission, spectral reflection and more. Signal processing theory learns that a classification of particles fundamentally improves as the number of independent features increases. At the laboratories of the Delft University a test bench is available for the development of advanced multi-sensor quality control and sorting systems. It is based on a software system for detection, segmentation and classification of particles, which was especially developed for this application. Work is now in progress for high-speed detection and sorting of non-ferrous alloys, including aluminium casting and wrought alloys. A recently developed actuator design enables positive sorting into more than 3 fractions of particles that are randomly orientated on a flat conveyor belt.