Automatic Detection and Early Warning of Safety Hazards in Complex Scenarios Using Graph Neural Networks

DOI

10.2991/978-94-6463-780-9_51How to use a DOI?

Keywords

Coal Mine Safety; Graph Neural Network; Multi-source Monitoring Data Fusion; Dynamic Time Convolution

Abstract

Due to the increasing complexity of coal mine production environment and the rapid growth of monitoring data, the traditional methods are difficult to timely and accurately detect potential hazard and take early warning measures. Starting from existing Graph Neural Networks (GNN) model, this paper innovatively and for the first time proposes to extract fine-grained features from multi-source monitoring data and downhole geological structure information collected by sensors through the fusion of multi-layer aggregators and dynamic time convolution modules. After that, the cross-graph attention mechanism is employed to realize the effective interaction and fusion of multi-channel data. Finally, the paper employs the depth graph representation learning to merge the spatial structure characteristics of the coal mine with the time series dynamic information to develop a multi-level visual risk map, which can result in accurately identifying the potential safety hazard and warning efficiently during the coal mine operation. The experimental results demonstrate that the accuracy of potential safety hazard identification of the proposed complex GNN model can be greatly improved compared to the existing methods.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Jingyu Wang
AU  - Cheng Xing
AU  - Heng Zhang
PY  - 2025
DA  - 2025/07/03
TI  - Automatic Detection and Early Warning of Safety Hazards in Complex Scenarios Using Graph Neural Networks
BT  - Proceedings of the 2025 International Conference on Engineering Management and Safety Engineering (EMSE 2025)
PB  - Atlantis Press
SP  - 558
EP  - 564
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-780-9_51
DO  - 10.2991/978-94-6463-780-9_51
ID  - Wang2025
ER  -