Search

Transport and separation processes of solid waste can only be modelled successfully with discrete element methods in case the shape of the particles can be described accurately. The existing techniques for morphological data acquisition, such as computed tomography, laser scanning technique, optical interferometer, stereo photography and structured light technique, are laborious and require a large amount of realistic solid waste samples. Therefore, there is a pressing need for an alternative method to describe the shape of solid waste particles and to generate multiple variations of particles with almost similar shapes. In this paper, a new method to describe solid waste particles is proposed that is frequency-based and uses spherical harmonics (SHs). Additionally, a new shape generation method is introduced that uses the shape description of a single particle to generate an array of related shapes based on a probability density function with a dimensionless control factor η. The newly proposed methods were successfully applied to describe the complex shapes of pieces of metal and plastic scrap. The shapes of these pieces of scrap can be described adequately with 15° of SH expansion and the overall divergence is within 0.1 mm. Five different values for η were tested, which generated shapes with the same distribution as the original particle. Rising levels of η cause the morphological variation of the generated particles to increase. These new methods improve the modelling of transportation and separation processes. [ABSTRACT FROM AUTHOR]

End-of-life vehicles (ELVs) provide a particularly potent source of supply for metals. Hence, the recycling and sorting techniques for ferrous and nonferrous metal scraps from ELVs significantly increase metal resource utilization. However, different kinds of nonferrous metal scraps, such as aluminium (Al) and copper (Cu), are not further automatically classified due to the lack of proper techniques. The purpose of this study is to propose an identification method for different nonferrous metal scraps, facilitate the further separation of nonferrous metal scraps, achieve better management of recycled metal resources and increase sustainability. A convolutional neural network (CNN) and SEEDS (superpixels extracted via energy-driven sampling) were adopted in this study. To build the classifier, 80 training images of randomly chosen Al and Cu scraps were taken, and some practical methods were proposed, including training patch generation with SEEDS, image data augmentation and automatic labelling methods for enormous training data. To obtain more accurate results, SEEDS was also used to optimize the coarse results obtained from the pretrained CNN model. Five indicators were adopted to evaluate the final identification results. Furthermore, 15 test samples concerning different classification environments were tested through the proposed model, and it performed well under all of the employed evaluation indexes, with an average precision of 0.98. The results demonstrate that the proposed model is robust for metal scrap identification, which can be expanded to a complex industrial environment, and it presents new possibilities for highly accurate automatic nonferrous metal scrap classification. [ABSTRACT FROM AUTHOR]