Fundamentals of Multi-Sensor Separation Systems
This paper gives an overview of the research at the Delft University of Technology in the field of automatic quality control and sorting systems, which are applied for solid bulk materials in the recycling industry. From theory it is known that a simultaneous detection of several features with sensors of different detection principles improves the quality of the classification dramatically. However, the realisation of such systems demands considerable engineering efforts before the benefits of multi-sensor detection and sorting can be explored. To accomplish this goal a universal data-processing system was developed. It was successfully tested with a prototype of a 2-sensor system for detection of CSP contamination in glass cullet at a rate of 200 particles per second and per meter sensor width.