Research on identification, separation and mechanism of soft and hard gangue from raw coal gangue via dual-energy X-ray based on machine learning.
Saved in:
| Title: | Research on identification, separation and mechanism of soft and hard gangue from raw coal gangue via dual-energy X-ray based on machine learning. |
|---|---|
| Authors: | Zhao, Haitao1 (AUTHOR), Ma, Xiaomin1,2,3 (AUTHOR) ma_xiaomin@126.com, Fan, Yuping1 (AUTHOR), Dong, Xianshu1 (AUTHOR), Kang, Yifeng1 (AUTHOR), Wen, Pengcheng1 (AUTHOR) |
| Source: | International Journal of Coal Preparation & Utilization. 2026, Vol. 46 Issue 6, p1713-1734. 22p. |
| Subject Terms: | *Machine learning, *Random forest algorithms, *X-ray absorption, *Coal mine waste, *Mineral analysis, *X-ray imaging |
| Abstract: | Traditional coal gangue often used in backfill, not conducive to bulk resource utilization, through the effective separation of raw coal gangue in the soft gangue and hard gangue, hard gangue bulk use in construction and other industries, soft gangue bulk use in the extraction of coal kaolinite recycling, etc. can be a highly efficient solution to the problem of bulk utilization of coal solid waste. In this paper, a method based on the combination of dual-energy X-ray and machine learning algorithms is proposed for the soft and hard gangue measurement, identification and separation of raw coal gangue. The results showed that the separation effect under the random forest classification model was optimal. In the randomized test set data, the soft and hard gangue yields of the gangue samples were 56.5% and 43.5%, respectively; the accuracies of the soft and hard gangue separation and identification were 91.15% and 86.2% respectively. The separation mechanism was analyzed via X-ray diffractometer (XRD), scanning electron microscope (SEM-EDS), BPMA-type automated mineral parameter analysis system (BPMA) and X-ray three-dimensional microscope (X-CT); additionally, the principle of dual-energy X-ray separation of soft and hard gangue was established on the basis of its different degrees of the X-ray attenuation. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Traditional coal gangue often used in backfill, not conducive to bulk resource utilization, through the effective separation of raw coal gangue in the soft gangue and hard gangue, hard gangue bulk use in construction and other industries, soft gangue bulk use in the extraction of coal kaolinite recycling, etc. can be a highly efficient solution to the problem of bulk utilization of coal solid waste. In this paper, a method based on the combination of dual-energy X-ray and machine learning algorithms is proposed for the soft and hard gangue measurement, identification and separation of raw coal gangue. The results showed that the separation effect under the random forest classification model was optimal. In the randomized test set data, the soft and hard gangue yields of the gangue samples were 56.5% and 43.5%, respectively; the accuracies of the soft and hard gangue separation and identification were 91.15% and 86.2% respectively. The separation mechanism was analyzed via X-ray diffractometer (XRD), scanning electron microscope (SEM-EDS), BPMA-type automated mineral parameter analysis system (BPMA) and X-ray three-dimensional microscope (X-CT); additionally, the principle of dual-energy X-ray separation of soft and hard gangue was established on the basis of its different degrees of the X-ray attenuation. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 19392699 |
| DOI: | 10.1080/19392699.2025.2505457 |
