
The results give mine operators concrete stress thresholds to prevent fault‑triggered collapses, improving safety and productivity in fault‑prone coal mines. Implementing the recommended monitoring and adaptive support can reduce costly roof failures and environmental impact.
Modern coal mining increasingly confronts the challenge of extracting thick seams intersected by steep faults. While traditional empirical rules offer limited guidance, advances in numerical modeling and physical simulation now enable engineers to map stress trajectories as excavation progresses. By integrating field measurements with similar‑material laboratory tests, researchers can visualize how vertical loading intensifies and horizontal stresses unload near fault planes, creating a dynamic pressure environment that traditional designs often overlook.
The Ordos Basin study quantified that floor stress can surge to nearly 20 MPa and roof stress can climb above 17 MPa once the working face is within five metres of a 70° normal fault. These spikes produce a distinctive U‑shaped subsidence curve punctuated by M‑shaped anomalies, signaling rapid roof‑failure cycles. Such precise thresholds empower mine planners to pre‑emptively reinforce critical zones, adjust mining sequences, and allocate resources where stress concentrations are predicted, thereby curbing unexpected roof falls and rock bursts.
Beyond immediate safety gains, the findings underscore a broader shift toward data‑driven risk management in underground mining. Deploying geotechnical sensors, real‑time stress monitoring networks, and predictive analytics creates an early‑warning ecosystem that can trigger automated support deployment or halt operations before fault activation escalates. As the industry embraces these technologies, the cost of downtime diminishes, regulatory compliance strengthens, and the environmental footprint of mining operations contracts, positioning firms to meet both economic and sustainability objectives.
Comments
Want to join the conversation?
Loading comments...