Machine Learning–Enhanced Microseismic Analysis for Evaluating Rock Crack Trajectory
Microseismic (MS) monitoring is an effective method for tracking the development of rock fractures. However, the utilization of existing data is severely limited by current visualization techniques. In this study, the evolution characteristics of MS parameters during the rock fracture process were i...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-01-01
|
| Series: | Shock and Vibration |
| Online Access: | http://dx.doi.org/10.1155/2024/6845665 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Microseismic (MS) monitoring is an effective method for tracking the development of rock fractures. However, the utilization of existing data is severely limited by current visualization techniques. In this study, the evolution characteristics of MS parameters during the rock fracture process were investigated using MS localization techniques and spatial clustering methods. Machine learning methods were applied to cluster acoustic emission events, enabling the automatic identification of hazardous regions and the construction of crack propagation trajectories. The applicability of clustering methods was assessed using MS monitoring data collected during an in situ rock fracture process. The research results show that the spatial clustering method can effectively delineate hazardous areas, and the changes in these regions reflect the progressive development of rock mass fractures. Regions with a higher proportion of high-energy MS events pose greater risks. Additionally, the propagation trajectory of MS events can effectively characterize the concentration and development direction of fractures. This study provides valuable insights into the characterization of crack evolution and seepage channels. |
|---|---|
| ISSN: | 1875-9203 |