Rockburst Prediction and Risk Assessment in Underground Engineering

Rockburst refers to the sudden and violent fracture of rock in high‐stress underground environments, posing a serious threat to excavations such as mines, tunnels and caverns. Prediction and risk assessment of rockburst events combine field monitoring, laboratory testing, numerical simulation and data‐driven analytics to anticipate occurrences and guide mitigation measures. Microseismic monitoring, stress measurement and geological mapping offer real‐time indicators of evolving rock mass conditions, while laboratory experiments and in situ testing elucidate the mechanical and thermal processes underlying rock failure. Advanced numerical models simulate stress redistribution around openings and predict zones of potential instability. In parallel, machine-learning and statistical techniques mine historical and live data to classify risk levels, providing probabilistic forecasts that inform support design, excavation sequencing and emergency planning. A holistic framework integrating physical understanding with data analytics is essential for robust risk management, reducing hazards to personnel and infrastructure in deep and complex underground projects globally.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Rockburst Prediction and Risk Assessment in Underground Engineering publication trend

The graph below shows the total number of articles in rockburst prediction and risk assessment in underground engineering across all publications each year (not limited to Nature Index journals).

Technical terms

Rockburst: Sudden, violent failure of rock mass under high stress conditions in subterranean excavations.

Microseismic monitoring: Recording of low‐magnitude seismic events to detect stress changes and fracturing within rock.

Principal component analysis (PCA): Statistical method that reduces data dimensionality by identifying main directions of variance.

Gradient boosting: Ensemble machine-learning approach that sequentially trains models to correct residual errors.

Artificial neural network (ANN): Computational system inspired by biological neurons, used for pattern recognition and classification.

In situ testing: Experimental measurements conducted directly within the underground environment to characterise rock behaviour.

References

1. Evaluation of Short-Term Rockburst Risk Severity Using Machine Learning Methods. Big Data and Cognitive Computing (2023).
2. Monitoring, Warning, and Control of Rockburst in Deep Metal Mines. Engineering (2017).
3. The Use of Data Mining Techniques in Rockburst Risk Assessment. Engineering (2017).