Abstract
ABSTRACT The wide range of materials and colors in waste plastic bottles makes the sorting process complex. This study investigates the reject rate of waste plastics using the GLP4 dual-path sorting machine, which utilizes both infrared and visible light. The recognition targets include blue and white PET bottles, green PET bottles, as well as other materials such as PC, PE and PP. Firstly, to optimize the air jetting process, computational fluid dynamics (CFD) analysis was utilized to simulate the impact of different nozzle diameters on the sorting efficiency. The results demonstrate that with a separator plate distance of 72 mm, the exclusion rate reaches 99.36%. Similarly, at an angle of 0°, the reject rate achieves 99.08%, while a nozzle diameter of 4 mm yields a reject rate of 99.58%. These findings validate the accuracy of our simulation results. Finally, by optimizing and adjusting HSV range values along with background saturation levels, we successfully reduce the omission proportion from 4.58% to 1.26% for PET white and blue plastic bottles; moreover, we decrease false recognition rates for white bottles made from PP (polypropylene), PC (polycarbonate) and PE (polyethylene) from 3.88% to 2.69%. Notably, there is no impact on false recognition rates for PE material which remains at an accurate recognition rate of 100%. For recognizing white, blue and green PET bottles specifically, the MSR color enhancement algorithm exhibits superior rejection performance with respective rejection rates reaching 94.8%, 95.83% and 90.29%. Overall, this research provides valuable insights for further development and application within waste plastic sorting.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call