- 1School of Engineering and Technology, China University of Geosciences (Beijing), 29 Xueyuan RD, Haidian Dist, Beijing 100083, China (xiaoyu.qi@email.cugb.edu.cn)
- 2Department of Civil and Environmental Engineering, Politecnico di Milano, Piazza Leonardo da Vinci 32, Milano 20133, Italy (stefano.mariani@polimi.it)
- 3School of Geological Engineering and Geomatics, Chang'an University, Middle-section of Nan'er Huan Road, Xi'an 710054, China (dicexy_1@chd.edu.cn)
Accurate identification and effective support of key blocks are crucial for ensuring the stability and safety of rock slopes. In previous studies, the number of structural planes and rock blocks was limited by considerations related to computational efficiency and capabilities, limiting the accurate characterization of complex rock slopes and hindering the identification of key blocks, potentially compromising stability and safety.
In this paper, a knowledge-data dually driven paradigm for accurate identification of key blocks in complex rock slopes is proposed. Our essential idea is to integrate key block theory into data-driven models based on finely characterized structural features to accurately identify key blocks in complex rock slopes. The proposed novel paradigm consists of (1) representing rock slopes as graph-structured data based on complex systems theory, (2) identifying key nodes in the graph-structured data using graph deep learning, and (3) mapping the key nodes of graph-structured data to corresponding key blocks in the rock slope.
Verification experiments and real-case applications were conducted using the proposed method. The verification results demonstrate excellent model performance, strong generalization capability, and effective classification results. The real case application is conducted on the northern slope of the Yanqianshan Iron Mine. The results show that:
(1) The proposed method has advantages in accurately representing the structural characteristics of complex rock slopes, which enhances the accuracy of key block identification;
(2) Integrating scientific knowledge of key block theory into GNNs facilitates the learning and capturing of internal structural characteristics of rock block systems and the distribution patterns of key blocks; and
(3) Our proposed paradigm is capable of accurately identifying key blocks from extremely imbalanced rock block systems, providing effective support and instability prevention of rock slopes.
How to cite: Qi, X., Meng, H., Xu, N., Mei, G., Peng, J., Mariani, S., and Vecchia, G. D.: A knowledge-data dually driven paradigm for accurate identification of key blocks in complex rock slopes, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-2926, https://doi.org/10.5194/egusphere-egu25-2926, 2025.
