Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal Mines
ABSTRACT Coal mine ventilation shafts are affected by a variety of factors such as ground stress, water intrusion, and corrosion due to prolonged use, leading to structural damage and deformation, which can adversely impact the ventilation efficiency and safety of the wind well. In response to the i...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-03-01
|
| Series: | Engineering Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/eng2.70083 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850207649734328320 |
|---|---|
| author | Tianji Lv Wenjun Fu |
| author_facet | Tianji Lv Wenjun Fu |
| author_sort | Tianji Lv |
| collection | DOAJ |
| description | ABSTRACT Coal mine ventilation shafts are affected by a variety of factors such as ground stress, water intrusion, and corrosion due to prolonged use, leading to structural damage and deformation, which can adversely impact the ventilation efficiency and safety of the wind well. In response to the issue of cracks in the walls of these vertical shafts, this study employs a comprehensive approach, integrating theoretical analysis, the construction of a robotic experimental platform, algorithm development, and field testing to design and research a safety inspection robot for detecting cracks in ventilation shaft walls of coal mines. A novel structure for a ventilation shaft inspection robot is designed. The robot's hardware system is developed, along with upper‐level computer software that enables visual monitoring and control. An improved YOLOv8n‐based network model is introduced, and the proposed crack detection algorithm undergoes a series of comparison and ablation experiments with different attention mechanisms and algorithm models. The experimental results show that the improved YOLOv8n model achieves a precision of 97.5%, a recall of 93.5%, and an average precision of 98%. The model size is only 1.63 M, with a computational load of 6.2 GFLOPs, and real‐time performance of 158.7 FPS, providing an efficient solution for detecting well‐wall cracks. |
| format | Article |
| id | doaj-art-3faff7b5c5cc4d49ae13cfbfe6addef1 |
| institution | OA Journals |
| issn | 2577-8196 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Wiley |
| record_format | Article |
| series | Engineering Reports |
| spelling | doaj-art-3faff7b5c5cc4d49ae13cfbfe6addef12025-08-20T02:10:27ZengWileyEngineering Reports2577-81962025-03-0173n/an/a10.1002/eng2.70083Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal MinesTianji Lv0Wenjun Fu1College of Safety Science and Engineering Liaoning Technical University Huludao ChinaBeijing China Coal Mine Engineering Co. Ltd. Beijing ChinaABSTRACT Coal mine ventilation shafts are affected by a variety of factors such as ground stress, water intrusion, and corrosion due to prolonged use, leading to structural damage and deformation, which can adversely impact the ventilation efficiency and safety of the wind well. In response to the issue of cracks in the walls of these vertical shafts, this study employs a comprehensive approach, integrating theoretical analysis, the construction of a robotic experimental platform, algorithm development, and field testing to design and research a safety inspection robot for detecting cracks in ventilation shaft walls of coal mines. A novel structure for a ventilation shaft inspection robot is designed. The robot's hardware system is developed, along with upper‐level computer software that enables visual monitoring and control. An improved YOLOv8n‐based network model is introduced, and the proposed crack detection algorithm undergoes a series of comparison and ablation experiments with different attention mechanisms and algorithm models. The experimental results show that the improved YOLOv8n model achieves a precision of 97.5%, a recall of 93.5%, and an average precision of 98%. The model size is only 1.63 M, with a computational load of 6.2 GFLOPs, and real‐time performance of 158.7 FPS, providing an efficient solution for detecting well‐wall cracks.https://doi.org/10.1002/eng2.70083detecting cracksinspection robotventilation shaftYOLOv8n |
| spellingShingle | Tianji Lv Wenjun Fu Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal Mines Engineering Reports detecting cracks inspection robot ventilation shaft YOLOv8n |
| title | Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal Mines |
| title_full | Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal Mines |
| title_fullStr | Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal Mines |
| title_full_unstemmed | Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal Mines |
| title_short | Design and Research of a Safety Inspection Robot for Detecting Cracks in Ventilation Shaft Walls of Coal Mines |
| title_sort | design and research of a safety inspection robot for detecting cracks in ventilation shaft walls of coal mines |
| topic | detecting cracks inspection robot ventilation shaft YOLOv8n |
| url | https://doi.org/10.1002/eng2.70083 |
| work_keys_str_mv | AT tianjilv designandresearchofasafetyinspectionrobotfordetectingcracksinventilationshaftwallsofcoalmines AT wenjunfu designandresearchofasafetyinspectionrobotfordetectingcracksinventilationshaftwallsofcoalmines |