PET flakes are delivered in BigBags and can have different mixtures. This leads to fluctuations in the production process, which can be reflected in the color value. The early evaluation of PET flakes and their quality prevents waste and costs. A partially automated control system for quality assessment should show how detection can be carried out.
In a furnace, PET flakes are melted in a bowl as a sample. A downstream camera system evaluates the PET flakes with a color camera in the LAB area. Using a mathematical model, PET flakes can be assigned to a quality group. The model calculates the proximity to a group based on the available data and improves with an increasing number of measurements.
The entire system is to be implemented offline on a laboratory scale near the big bag station. A conventional PC can be used for the software.