收藏切换
Research on Source Identification and Migration Law of Slag-Type Debris Flow Based on Deep Learning
收藏切换
Lei ZHANG1, 2, 3, Ao ZHENG1, Shangkai HAO4, Jiayuan WANG1, Qianru YAN1, Zhipeng SUN1
Mining Research and Development | 2026, 46(2) : 167 - 174
Less
收藏切换
Mining Research and Development | 2026, 46(2): 167-174
Research on Source Identification and Migration Law of Slag-Type Debris Flow Based on Deep Learning
Full
Lei ZHANG1, 2, 3, Ao ZHENG1, Shangkai HAO4, Jiayuan WANG1, Qianru YAN1, Zhipeng SUN1
Affiliations
  • 1.College of Coal Engineering, Shanxi Datong University, Datong, Shanxi 037009, China
  • 2.Shanxi Provincial Key Laboratory Cultivation Base for Coal Mine Water Jet Technology and Equipment Co-built by the Department and the City, Datong, Shanxi 037009, China
  • 3.Shanxi Provincial Technology Innovation Center for Mining Area Geo-environmental Monitoring and Emergency Disaster Reduction, Datong, Shanxi 037009, China
  • 4.Shanxi No.3 Institute of Geological Engineering Exploration Co., Ltd., Jinzhong, Shanxi 030602, China
Published: 2026-02-25
Outline
收藏切换

Aiming at the limitations of the current debris flow source detection algorithm, a YOLOv5s-GCE model was proposed. This model introduces CBAM attention mechanism, Ghost convolution and EIOU bounding box loss function to improve its identification accuracy and efficiency. In order to verify the detection effect of YOLOv5s-GCE model, it was compared with YOLOv4 and YOLOv5s models. The results show that the average accuracy of the YOLOv5s-GCE model is 95.4%, which is 2.5 and 8.6 percentage points higher than that of the YOLOv4 and YOLOv5s models, respectively. The detection accuracy of YOLOv5s-GCE model for debris flow source is as high as 96%, and the detection effect is better. The particle flow numerical simulation method was used to study the migration law of debris flow. The simulation results show that the front end of the source body is destroyed and migrated in the initial stage, and the infiltration of water particles leads to the increase of porosity and structural instability of the source body. The mid-section of the source body is subjected to compression from the rear end and restraint from the front end, leading to a significant reduction in porosity, structural densification and the occurrence of localized bulging. In the later stage of migration, a large number of material sources accumulated at the gully mouth. The threat of debris flow to downstream construction facilities is significant, and the soil on the left side of the mid-section of the source area is the most threatening in extreme weather. The research results are of great significance to the prevention and control of slag-type debris flow.

Slag-type debris flow  /  Source identification  /  Target detection  /  Migration law  /  Numerical simulation
张磊, 郑奥, 郝尚凯, 王佳源, 燕倩如, 孙志鹏. 基于深度学习的矿渣型泥石流物源识别及运移规律研究. 矿业研究与开发, 2026 , 46 (2) : 167 -174 .
Lei ZHANG, Ao ZHENG, Shangkai HAO, Jiayuan WANG, Qianru YAN, Zhipeng SUN. Research on Source Identification and Migration Law of Slag-Type Debris Flow Based on Deep Learning[J]. Mining Research and Development, 2026 , 46 (2) : 167 -174 .
Year 2026 volume 46 Issue 2
34
9
Cite this Article
BibTeX
Article Info
  • Receive Date:2025-01-22
  • Online Date:2026-10-29
  • Published:2026-02-25
Article Data
Affiliations
History
  • Received:2025-01-22
Funding
Affiliations
    1.College of Coal Engineering, Shanxi Datong University, Datong, Shanxi 037009, China
    2.Shanxi Provincial Key Laboratory Cultivation Base for Coal Mine Water Jet Technology and Equipment Co-built by the Department and the City, Datong, Shanxi 037009, China
    3.Shanxi Provincial Technology Innovation Center for Mining Area Geo-environmental Monitoring and Emergency Disaster Reduction, Datong, Shanxi 037009, China
    4.Shanxi No.3 Institute of Geological Engineering Exploration Co., Ltd., Jinzhong, Shanxi 030602, China
References
Share
https://castjournals.cast.org.cn/joweb/kyyjykf/EN/
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT